When the US Securities and Exchange Commission (SEC) finalized a rule that made major changes to US equities market infrastructure in December 2020, it opened the door for interested technology vendors to become suppliers of consolidated market data.

The new rules create a system in which, instead of two exchange-run securities information processors (Sips) pumping out bid/ask quotes consolidated from US trading venues to consumers, a decentralized system of entities called competing consolidators will perform that role.

Some technology vendors with experience in market infrastructure have expressed interest in becoming competing consolidators, including MayStreet, McKay Brothers, Activ Financial, NovaSparks and the Miami International Securities Exchange (Miax).

These vendors believe they have unique capabilities and experience they can leverage to become the new Sips. However, although they will be providing an industry service, those services will be operated as commercial solutions rather than as utilities, and that exchange data doesn’t come for free. So before they can formulate the business plans that will underpin their bids, they need to know how much the exchanges will charge them for the data they will be using, and how revenues will be distributed.

“Everything revolves around how high the new Sip fees redistributed to the exchanges will be. That is the parameter that will make or break this modernization initiative,” says Stephan Tyc, co-founder of McKay Brothers and Quincy Data.

The SEC hopes that competing consolidators will offer customers a range of products that use consolidated data from the exchanges, and in parallel, has also expanded the definition of what that data is. Currently, the Sips include top-of-book quotes. Under the new dispensation, competing consolidators will be able to offer depth-of-book data, auctions, odd lots, and regulatory data.

Tyc says the competing consolidators will only be able to determine the price at which they sell these products once they know what the fee schedule under a new plan for governance of the Sips looks like.

“The price will come on top of the fees that the governance committee decides to redistribute to the exchanges. So the competing consolidators must live on the extra margin they make on the price of the Sip, minus the cost of data that is the input needed to create the Sip,” he says. 

Regulatory uncertainty

So where is this all-important fee schedule?

In January 2020, the SEC published a proposed order that would oblige the Financial Industry Regulatory Authority (Finra) and the exchanges—also referred to as the self-regulatory organizations (SROs)—to draw up a new plan for governance of the Sips. The governance of the Sips, the SEC says, is marked by inherent conflicts of interest, as the exchange groups that sit on the operating committees of the Sips have disproportionate voting power compared to the market participant representatives. The SEC ordered that the new plan would give more voting power to the broker-dealers and investment firms on these committees, among other changes.

The SROs drafted their new plan and filed it with the commission in October 2020. In their notice of filing, they said that once the governance plan was approved, the new operating committee it created could decide on the fees they would charge for data. Those fees would be included in amendments to the approved governance plan.

The approval of this plan is the first key milestone on the road to the creation of competing consolidators. But the SEC did not approve it, or even deny it, but decided rather to send it out for public comment in January 2021—a move that has introduced substantial confusion into the timeline of the Sips’ modernization efforts, according to the SROs.

And to compound the uncertainty, Nasdaq and the New York Stock Exchange (NYSE), which administer the two current Sips, have brought court action against the SEC to halt the infrastructure rule, which they see as an existential threat to their businesses. The exchanges and their supporters say the Sips have served the market well for decades, proved resilient during the extreme volatility of the March 2020 financial crisis, and the exchanges have spent considerable resources improving their speed to reduce data latency. Why mess with a good thing? In February, the two exchange groups made parallel filings in the US Court of Appeals for the District of Columbia Circuit to review the SEC’s infrastructure plan.

The SROs also want the SEC to stay the infrastructure rule until the new governance plan is approved, saying there’s too much confusion over timelines otherwise.

Vendors undeterred

The competing consolidator hopefuls would be forgiven for feeling daunted by all this uncertainty. As consultant Bill Harts tweeted on February 9, “… The incentives for companies to create competing Sips are rapidly eroding. Who’s going to commit investment [dollars] to build now when it could be worthless at the end of this suit?”

However, some of these hopefuls tell WatersTechnology that they remain unfazed. For one, they say, most observers expected the SROs to take the SEC to court, as various exchanges have warned as much in comment letters, and because they have a history of challenging the SEC successfully in court. These vendors say they have been factoring in the possibility of legal delays.

Another reason, they say, is that the SEC has planned a long timeline of implementation of the infrastructure rule—it will take, at the very least, two years to get to the point where competing consolidators are built, vetted, and approved to operate Sips.

And thirdly, these vendors say, they fully expect the infrastructure rule to go ahead because it has wide support outside of the SROs. The idea of competing consolidators has been kicked around in various forums by the industry and the SEC; the SEC itself signaled in 2012, under a Democratic administration, that it was considering modernization of the National Market System. The SEC’s Trump-era pick, former chairman Jay Clayton, and former director of the Division of Trading and Markets Brett Redfearn, both champions of these rules, have recently left the agency, but there is no reason to think the next chairman, Gary Gensler, won’t be just as enthusiastic about adopting it. 

“We were anticipating that this [legal action] would happen. And I think there is bipartisan support for these proposals. The court action can only slow the process down, but even this is not certain: The SEC decided on an implementation schedule that was quite slow and cautious, which was a very good decision, as it gives them time to fight” the SROs in court, McKay’s Tyc says.

Tyc says McKay would be a good contender as a competing consolidator because of its experience as a telecoms company, providing extreme low-latency microwave bandwidth for trading firms.

“McKay has the fastest networks between the three centers in New Jersey [Carteret, Mahwah, and Secaucus]. So for those people who want the fastest Sip to guarantee their best execution requirements and do all the other things they need with the mostly timely data, we believe we can produce that, because we can ferry around the data much faster than others. When I say much faster, I mean about one microsecond faster—but that counts,” he says.

For NovaSparks, another company that’s interested in becoming a competing consolidator, the delays and uncertainty are not a major headache, as what the company would offer as a Sip provider is a solution it is working on anyway, says CEO Luc Burgun.  

NovaSparks specializes in field-programmable gate array (FPGA) ticker plant and feed handler appliances for ultra-low latency applications. It plans to introduce an FPGA-based solution for market data consolidation to support the new rule featuring the lowest possible latency. FPGA is firmware, a programmable semiconductor.

NovaSparks has been working on this solution for the past two years. “What we are developing today is something we need regardless of the new Sips. There happens to be a convergence between our roadmap and new Sips roadmap, which is perfect for us. For the time being, we don’t have to develop anything specifically for the current and future Sips,” Burgun says.

And it’s possible to make general estimations of what kinds of services the exchanges will want to charge competing consolidators for.

MayStreet CEO Patrick Flannery did not want to comment directly on the court action, as it is still ongoing. However, he tells WatersTechnology that he expects the exchanges to charge at least an access fee for the data the new Sips will take in.

“They are going to have to pay for cross-connections, and/or other co-location telecoms, and they are going to have to pay, we assume, some sort of access fee to receive the data. It seems unlikely that the exchanges will not consider charging for these sorts of things,” Flannery says.

The competing consolidators could be charged based on their market share over a particular period, he adds. “It could be flat fees, which could get quite large, so that would be a problem. The exchanges will probably argue that they want both—they want a fixed fee and revenue share flat fee, because that will be most advantageous. But we will see what the SEC ultimately approves,” he says.

MayStreet has a platform that allows firms to manage their market data; this is the core of the business, Flannery says. The competing consolidators are going to have to be scalable and reliable, he adds.

“We have software today that builds consolidated feeds from a number of sources globally. The consolidation we do in US equities, where we take proprietary feeds and build a comprehensive US equity order book, is very similar [to building a competing consolidator]. The inputs for competing consolidators would be different, but we have a lot of the software and logic and expertise,” Flannery says.

Shane Swanson of consultancy Greenwich Associates says MayStreet would be an obvious choice as a contender. “They have a lot of experience; they run the Market Information Data Analytics System (Midas) platform for the SEC. This is well within their ambit,” he says.

MayStreet announced in 2019 that it had assumed the role of market data provider for Midas, after acquiring part of the business of the previous provider, Thesys Technologies. Midas helps the SEC monitor and understand market trends.

McKay and NovaSparks, on the other hand, will have an advantage in that they are focused on the low-latency space, Swanson says. “Reading data, passing it along, and publishing it is not a huge step change [for these providers], although there are certainly differences and challenges in being part of a quasi-regulatory engagement,” he says.

Miax, which has also signaled its interest, has an advantage in that it is an exchange, and therefore already compliant with Regulation Systems Compliance and Integrity (Reg SCI), a set of rules that monitor the security and capacity of exchanges, alternative trading systems, and other market infrastructure. Competing consolidators will be subject to the rule.

“Reg SCI is no concern for [Miax] as they are already compliant. I’m not sure that gives them a huge advantage, but it certainly provides them with the comfort of not having to do a huge upgrade to be compliant with Reg SCI,” Swanson says.

WatersTechnology reached out to Miami Exchange Group and Activ Financial for this article. Miax declined to comment, while Activ did not respond to requests for comment.

What the future holds

Exactly how the Sips will look when they are run by competing consolidators is difficult to say. The SEC hopes that each competing consolidator would offer its own distinct products and services to market participants, in contrast with the Sips, which are quite a uniform product. This, in theory at least, would introduce competition and innovation to the space.

“The idea is interesting in theory, and it’s good to hear that vendors are throwing their hats into the ring. I don’t know how well it would work in practice—no one does, because it has never been done! But I think it could be workable if the competing consolidators get enough volume to be profitable, because there are some clear benefits to having competition in this space,” Swanson says.

One point of speculation is how many competing consolidators would be optimal. The SEC says any more than 12 would be unwieldy. Flannery, on the other hand, says, “We do not necessarily need five good solutions, but we do need one good solution.”  

“Twelve feels like it would be too many given the potential for added complexity, although there are more than 12 equity exchanges, so it could be viable,” says Flannery. “To get the benefits of the [competing consolidator] model, the steady state would probably be three, but that is a total guess about products that haven’t even been deployed yet.”

Running a utility is as much an administrative problem as it is a technology problem, he adds. “You have to onboard customers quickly, report all the details, monitor, set controls—all these aspects of running the service are potentially more substantial than the technology challenges on their own,” he says. 

McKay’s Tyc says competing consolidators already exist, in that banks already consume market data from the exchanges’ proprietary feeds, and use it to assemble their own internal Sips. These firms are obliged to create their own centralized books and their own market data, partly because it’s just more efficient, and partly because they are required to execute client orders with the best data available.

“So those people who say that many Sips will be created are missing the point: There are already many, many Sips. The difference will be that those Sips, which are difficult to create, will be something you can buy instead of something you have to assemble yourself. It should make things a lot easier for people that do client order execution: They could choose to be self-aggregators and continue doing business as they did before, or they could buy a Sip from a competing consolidator,” Tyc says.

One advantage is that because each competing consolidator will be vetted by the SEC, individual banks’ compliance departments will not need to approve all the Sip developments in house, Tyc says. Obtaining those bank approvals can be “a big headache,” he adds.

Another advantage of having vendors offering new products is that consumers will have more choice, Swanson says.

“If you could use a competing consolidator, you could say, ‘OK, who is faster? Who has technology that is more in line with my own technology?’ Maybe they offer a binary protocol, and you use binary, and that is better for your methodology. Maybe you care about price, maybe you care about stability, maybe one vendor is fast but has more failures, and that is or is not OK for you and your needs. So it becomes a vendor relationship, rather than a mandatory one” like the current Sips], he says.

For MayStreet, fee models are an area where competing consolidators could provide innovation. The new Sips will be able to charge what they like, within the constraints of the market, but there should be a clear pricing model, he says.

Currently, Flannery says, there are two Sips that feed two, separate consolidated tapes of exchange data—the Consolidated Tape Association, run by NYSE, and the Unlisted Trading Privileges (UTP) Plan, run by Nasdaq—and users have to pay for two sets of connectivity: cross-connect and co-location fees, as well as access and redistribution fees. And then there are redistribution fees, which he says are capped at just under $1.9 million per month. And then there are display and non-display fees and professional and non-professional licenses.

“We want to get to where there are no distinctions between professional and non-professional, display and non-display. We think it would be optimal for the competing consolidators that offer one feed of consolidated US equity market data, not Tape A, Tape B, Tape C. That could help lower the connection charges, since some of the cross-connects are quite expensive. Having flat, transparent, per-cross-connect fees would make it more of a pay-per-use model,” Flannery says.

If billing models were simpler, there would be less need for audit rights, he adds. Getting rid of those would represent another saving, as it costs money to keep track of data usage.

“So where we would like to get to is a much more reasonable fee for cross-connect, and then a small, medium, and large redistribution fee. The really large retail brokers would pay a fraction of what they are currently paying if they are at or near the [redistribution] cap,” Flannery says.

Many of these vendors say they will probably partner with others to deliver their Sips as competing consolidators.

Burgun says the current Sips are “une usine a gaz” (literally, a gas factory)—a French expression meaning something overly complex and unwieldy. They aggregate huge amounts of data from over 16 different equities exchanges, extract the top-of-book, and send the feeds out, and they do all that with software, requiring large server farms.

“The new Sip is going to be even worse because they are going to add odd lots, order book, auctions—it’s going to be massive. Plus market data volumes are growing, as there are more and more registered stock exchanges,” he says. A competing consolidator using an FPGA solution could be faster and more reliable than software, and take up less space.

The FPGA solution could be co-located with a user’s direct feed in Secaucus, NJ, for example, and compute the Sip there; it wouldn’t have to only be in Mahwah or Carteret, where the NYSE and Nasdaq Sips are located, respectively, Burgun says. “The solution could be replicated anywhere at low cost, and everybody could get access to the Sip and with a pretty good latency profile in comparison with the direct feeds.” 

However, he adds, NovaSparks would not be able to deliver its FPGA solution alone, and would need to partner with another company for infrastructure and managed services. The company is already looking for such a partner as part of the work in developing its solution.

“I don’t believe that a tool solution provider can make it alone,” Burgun says.

He says the other competing consolidator contenders have some capabilities in terms of managed services, but, apart from McKay, don’t have wireless network capabilities.

“And they don’t have strong FPGA tools either, which means that if they do this with software, their aggregation is going to be based on a large PC farm again, and it will be difficult to deploy. For NovaSparks, the ideal would be to partner with a good IT network infrastructure provider and a good managed services provider,” Burgun says.

One of the criticisms of the Sips is that the current plans have created a two-tier system in which, in one memorable metaphor, clients who rely on the Sips are driving down a rutted, potholed country road, while those who can afford the exchanges’ proprietary feeds are hurtling down a six-lane highway. The validity of this metaphor aside, Burgun says that one possible innovation of the competing consolidator regime could be in creating a third, middle tier.

“Maybe there is a way to create a better Sip that is going to fulfil the needs of those who are fine with a low-end solution because they don’t want to pay any more than they do already. In the middle-end segment, you could get access to more market data (order book depth, odd lots) if you paid a bit more.” 

Unlike the efforts to establish an industry-led consolidated tape in Europe, which ran into a roadblock when no companies volunteered to run the project, the new Sip regime has the distinct advantage of already having companies willing and able to build a solution that meets the SEC’s requirements. In this case, the roadblock is the exchanges themselves blocking reform, and the success or failure of the SEC’s plans may hinge on the courts’ decisions.