Industry slams market data fee filing

In early 2020, the US Securities and Exchange Commission (SEC) kicked off a project to modernize securities information processors (Sips), the entities that pump out national markets system quote data to consumers. The current Sips, governed by the CTA/CQ and UTP plans, include top-of-book data in their feeds; in the commission’s new and improved world—made a reality by the Market Data Infrastructure (MDI) rule—that “core data” was expanded to also include depth-of-book and auctions information. This data would be disseminated by a new kind of processor: competing consolidators that would bring competition and product differentiation to the public feeds.

Since the MDI rule was passed in December 2020, the current plans had to come up with a plan for how the exchanges would charge the competing consolidators and others for this newly expanded core data. That was due by November 5 of last year. Potential competing consolidators and market data end-users were watching anxiously to see what these proposed fee schedules would do—perhaps these exchanges would offer raw ingredient market data at a fixed price for competing consolidators, and waive redistribution fees for these new Sips. At any rate, would the large exchanges present something that would make becoming a consolidator an attractive proposition?

Judging by the comments sent to the SEC on the proposed fees (the comment period ended on December 17), the answer is a resounding no.

This won’t be a surprise to readers of this column, who will recall that when the fee schedule was released, I interviewed Luc Burgun, chief executive officer of NovaSparks, a vendor that is preparing to register as a competing consolidator. Burgun said at the time that he believed the filing would make Sip top-of-book data about as expensive as the exchanges’ proprietary feeds, even though the product was less valuable as it would contain less depth and insight. So, in that case, why would any user buy the Sip data?  

Not to mention the fact that the plans proposed charging redistribution and access fees, as well as failing to consider that the exchanges would no longer bear the expense of the processors (that would be passed on to consumers via competing consolidator fees).

Almost all of the 14 comment letters on the SEC’s site are from market participants and their lobbyists, competing consolidator aspirants, and smaller exchanges, and all took issue with the fee filings for these reasons.

Firstly, the commenters weren’t happy with the submitting exchanges’ calculation of depth-of-book charges. Under the MDI rule, core data must include depth-of-book information at five aggregated price levels, expanded from just top-of-book quotes. The large exchanges started from the premise that they would charge based on the value of this data rather than the cost, as the plans have little information about the competing consolidators’ technology and connectivity, and therefore have no idea how much it would cost the new Sips to redistribute it. With that in mind, they compared prices for three proprietary top-of-book data products to the prices for three depth-of-book products by the same exchanges, arriving at an average ratio of 3.4. Then they used this ratio to multiply Sip fees to set the per-user fees of new depth-of-book data.

The result, Burgun and others say, is that the Sip feeds wind up being less attractive than the exchanges’ prop feeds. And these products aren’t comparable anyway, since the proprietary feeds used in the calculation include top-of-book information while the exchanges are proposing to set fees for data that does not include top-of-book. A cost-based, rather than a value-based approach would have been preferable, these critics say.

Fees charged by monopolistic providers such as the exclusive Sips need to be tied to some type of cost-based standard

Securities Industry and Financial Markets Association comment letter

“Fees charged by monopolistic providers such as the exclusive Sips need to be tied to some type of cost-based standard in order to preclude excessive profits if fees are too high, or underfunding or subsidization if fees are too low,” the Securities Industry and Financial Markets Association’s letter says.

MayStreet is another competing consolidator hopeful that submitted a comment letter. I spoke to its chief policy officer, Manisha Kimmel, for the Waters Wavelength podcast recently. Kimmel says the controversy around the fee schedule gets to an interesting question around what the cost of the underlying content is versus the cost of creating that content and whether it is divorced from the distribution of that content.

“The way the rule is written is that the SROs could just give proprietary datafeeds to the competing consolidators, and they would figure out how to get the levels of depth and create the NBBO—that would be on them. With that sort of world, what would be the true cost to the exchanges? It’s basically very incremental costs, because they already have this infrastructure around their proprietary datafeeds,” Kimmel says.

“And if you extend that concept further then you would expect a fixed cost for the data, because it doesn’t cost the exchange more if they’re supplying that data and it’s being distributed to two versus 2,000 people.”

If you were to open a bakery, Kimmel says, why would your sugar supplier charge you differently depending on whether you made fancy cupcakes or simple ones?

Also, she says, the proposed fees treat competing consolidators as if they were market data vendors. The commission itself says in the infrastructure rule that the competing consolidators are to be treated as Sips, not as vendors, which means they are exempt, for instance, from redistribution fees.

MayStreet CEO Patrick Flannery says in the company’s comment letter that the competing consolidators would consume “underlying content” while producing the consolidation product, which makes them generators of market data—this confusion between the raw material and the finished product is at the heart of the company’s concerns with the plans.

Kimmel explains that access and redistribution fees have nothing to do with the value of the underlying content.

“There was no change [in the fee filings] to the contracts for vendors. And it was stated that vendors and competing consolidators would have the same contracts. But how can that be the case when you switch to selling a different product—underlying content versus consolidated market data—and you’re selling to different people? You’re selling these raw ingredients to competing consolidators and self-aggregators, and only competing consolidators can distribute that data,” she says.

Some of the commenters—notably the smaller exchanges—call for the fee-setting responsibility to be reassigned to the CT Plan operating committee. The CT Plan is the scheme for the governance of the new market data system that the commission created in parallel with the MDI rule. Its implementation is currently held up by litigation—the large exchanges are seeking to have the courts vacate the rule. The CT Plan gives more votes on its operating committee to non-SROs by assigning votes per exchange group rather than per individual exchange. The CT Plan might never make it into existence; even if it does, it will be delayed. This delay is preferable, says John Ramsay, chief policy officer at IEX Group, to a captured and unfair voting process.

It was stated that vendors and competing consolidators would have the same contracts. But how can that be the case when you switch to selling a different product—underlying content versus consolidated market data—and you’re selling to different people?

Manisha Kimmel, MayStreet

“Because the formation of that operating committee and the operation of the CT Plan depends on resolution of pending litigation, progress on a fee structure that meets the commission’s public policy goals would regrettably be deferred. But at least the setting of fees would benefit from consideration by a broader set of voting stakeholders and a fairer and less conflicted governance structure. This latest unfortunate experience shows compellingly how badly those governance changes are needed,” Ramsay concludes.

The letters mostly agree that the fee filings fly in the face of what the commission mandated them to achieve—to expand access to market data and level the playing field between proprietary datafeeds and the public tape—for the sake of protecting those lucrative prop feed businesses.

The exchanges, on the other hand, have always maintained that the MDI rule was flawed from the start. Why modernize the Sips when they are modern already—a reliable, consolidated feed of top-of-book data? Those consumers that need depth can obtain it from the prop feeds, which are priced competitively and fairly and aren’t set by some rate-making authority, they say.

Nasdaq submitted the lone letter in support of the fee filings. In it, vice president Erika Moore says the prices in the fee filing, having been based on a methodology that referred to existing products, were fair and shaped by market forces.

“The NMS plans have historically used the value of the data to consumers to set fees,” Moore writes. “Professionals pay higher fees than Main Street retail investors, and algorithms, dark pools, and electronic traders pay higher fees than human professionals. This is an efficient—and fair—allocation because professionals earn more from the data than retail, and algorithms, dark pools, and electronic traders earn more than human professionals.”

The SEC has offered no guidance on how a cost-based rate system might work in practice, she says. A cost-based structure would be overly elaborate as it would require the operating committee to understand what all the different exchanges charge for different products. Plus, it’s contrary to the commission’s Congressional mandate, which is to rely on competition wherever possible in setting the price of NMS data.

On the charge that the large exchanges are saving money by shifting the costs for the Sips, the Nasdaq letter says the MDI rule requires the current Sips to continue operating for a while, so the exchanges aren’t saving anything right now. “Moreover, we believe that the impact of any such future adjustment on the fees would be immaterial,” it adds.

The SEC has a couple of months to either approve or disapprove the fee proposal. Obviously, which side of this debate you fall on will depend on who you are and where your interest lies, but I can’t see how even the most biased observer can expect the commission to approve the fee filing proposal in its current state.

Even if you agree with the exchanges—that the Sips are modern already, that the fee schedule is perfectly fair and reflects the most efficient, market-driven pricing estimates possible—the proposal is simply not compatible with the new world where there are competing consolidators disseminating NMS data, and the commission is apparently determined to get to that world. The businesses that would be registering as those competing consolidators are telling the commission loud and clear that this fee schedule is unworkable.