Today, the Commission held that neither NYSE Arca, Inc. ("Arca") nor Nasdaq Stock Market LLC ("Nasdaq") had met its burden to show that the fees, filed in 2010 for ArcaBook and Level 2 (together, the "products") and subsequently challenged by SIFMA, were fair and reasonable under the Securities and Exchange Act of 1934.
While we voted to support this decision, we write separately to (1) explain our rationale and note the limited precedential effect of this holding, (2) indicate the type of analysis we would find useful in determining whether a fee filed for an exchange's non-core product or service is fair and reasonable, and (3) raise a critical policy question underlying these proceedings as well as the pending challenges to other exchange fees.
I. We Concur.
In this matter, the Commission applied a market-based test to evaluate whether the exchanges' fees for the products were fair and reasonable. Each exchange was required to show, by a preponderance of the evidence, that it was "subject to significant competitive forces in setting the terms of its proposal for non-core data, including the level of any fees." Informed by the D.C. Circuit's analysis in NetCoalition v. SEC, our inquiry focused on two competitive forces: each exchange's need to attract order flow and the availability of alternatives to each product.
We agree with our colleagues on the Commission that, on the record before us, the exchanges failed to meet their respective burdens to show either that (1) an increase in the price of the product causes a loss of order flow, or (2) an increase in the cost of the product causes customers to utilize substitutes for the products offered by the respective exchanges. Moreover, the exchanges did not ask us to sustain their fees on the basis of cost.
To us, the exchanges' failure begins with the fact that customers are described as one undifferentiated group, despite the fact that market data products are typically priced with particular customers in mind (e.g. display vs. non-display, professional users vs. non-professional). The Commission's opinion necessarily analyzes evidence in the record developed by the parties. We would have liked to explore the differing levels of demand among these particular customer groups as well as the costs of switching to any alternatives that may be available to them. The exchanges, however, presented an oversimplified analysis, foreclosing evaluation of facts that may have been indicia of competitive forces that these distinct customer groups exert.
The exchanges' other failure was their cursory presentation of the platform theory — the theory that exchanges compete with each other, and with non-lit venues, through total cost rather than particular fees. Neither exchange sufficiently established that the total cost of trading on its platform (including transaction fees, access fees, market data fees, and other fees) might constrain its fees with respect to the products at issue or depth-of-book data more generally. Consequently, we are unable to determine, based on the evidence on the record, that platform-level competition exerts disciplinary power on the exchanges.
While we believe the exchanges' argumentation and evidence failed to prove competition in the market for depth-of-book data generally, we emphasize that the evidence on the record also does not establish a lack of competition in that market. Moreover, we do not believe this failure necessitates a conclusion that either exchange has market power with respect to the specific products at issue in this proceeding. The statistical evidence in the record, which suggested that demand may be inelastic for the products, was inconclusive as to market power: this apparent inelasticity may show market power, but it may also show that (regardless of whether the exchanges have market power) these products are so valuable that customers would be willing to pay much more for them before switching to an alternative, even in a competitive market.
In sum, we believe the Commission's holding in this case should be read only as dispositive of the record presented — not as a conclusive statement about the market for the products in question or the market for depth-of-book data more broadly.
II. What is Fair and Reasonable?
Since the inception of this matter, over four hundred rule changes by exchanges, and plan amendments filed by participants in National Market System plans, have been challenged as similarly unjustified. The Commission has remanded all of those challenges to the respective exchanges to develop or identify procedures to review them as potential denials or limitations of access to services, apply their procedures to the challenges, and submit to the Commission a record explaining their ultimate conclusions. The Commission's opinion gives little guidance as to what standards or analysis the exchanges should consider as they undertake this momentous task.
In our view, the exchanges have at least three paths forward. One would involve an analysis that shows how they segment the market for each product or service based on the types of customers it will serve, and then an assessment of how these different types of customers use and value such products or services. This approach could permit an analysis of what alternatives are available to these customer groups and illuminate how (or if) such customers can exert disciplining pressure on the exchange's pricing for that product or service. Another option would be for the exchanges to fully assess the platform theory and identify how customers who pay for a particular product or service exert competitive pressure on an exchange as a platform. Both analyses are undeniably complex, but such diligence is necessary to ensure that customers are not paying fees so high that they amount to denials of access. A third option is for the exchanges to justify their fees based on the costs to develop and offer their products or services. This approach would involve the most intervention and monitoring on behalf of the Commission, which would essentially be required to step into the role of rate-setter. From our perspective, this is the least appealing outcome.
III. For Our Part . . .
Through footnotes and parentheticals, the Commission's opinion hints at an important point: the Order Protection Rule motivates demand for these and other proprietary market data products and may make it more difficult for firms to exert market pressure on exchanges by diverting order flow elsewhere. We now add our voices to the chorus of commenters, including former Commissioners, who have raised concerns and questions about the Order Protection Rule.
With the Order Protection Rule, the Commission has dulled the market's normal disciplining forces, setting in motion a cascade of unintended consequences. It is not clear to us whether, in today's complex and high-tech market (where trading occurs in multiple types of venues), this rule's benefits outweigh the immense costs it generates. These costs will only grow as market participants spend significant sums litigating each increase in data fees, which will almost certainly involve expensive analyses to establish that each increase is fair and reasonable under the Exchange Act. We doubt whether the process that our decision seems to require is sustainable for exchanges, other market participants, or the Commission, all of whom will have to commit resources to developing and reviewing even more lengthy fee justifications. We believe that the Commission is ill-suited to the role of rate-setter for these, or any other, markets. It appears that Congress agreed with this sentiment when it made self-regulatory fee filings immediately effective, pursuant to Exchange Act Section 19(b)(3).
Accordingly, we call upon the Commission to do a retrospective review of the Order Protection Rule and other interrelated aspects of equity market structure: Why do we still have the Order Protection Rule? Who are we protecting? From what? And would other rules or market dynamics serve to protect these interests in a more efficient, and less costly, manner?
 In the Matter of the Application of Securities Industry and Financial Markets Ass'n For Review of Action taken by NYSE Arca, Inc. and Nasdaq Stock Market LLC, Exchange Act Rel. No. 84432 (Oct. 16, 2018); see also Exchange Act Section 3(a)(22)(B), 15 U.S.C. §78c(a)(22)(B); Exchange Act Section 11A(c)(1)(C) & (D), 15 U.S.C. § 78k-1(c)(1)(C) & (D); see also Rule 603(a) of Regulation NMS, 17 C.F.R. § 242.603(a).
 Order Setting Aside Action by Delegated Authority and Approving Proposed Rule Change Relating to NYSE Arca Data, Exchange Act Release No. 59039 (Dec. 2, 2008), 73 Fed. Reg. 74,781 (Dec. 9, 2008), available at https://www.gpo.gov/fdsys/pkg/FR-2008-12-09/pdf/E8-28908.pdf ("2008 ArcaBook Approval Order").
 615 F.3d 525 (D.C. Cir. 2010).
 In the Matter of the Application of Securities Industry and Financial Markets Ass'n and Bloomberg L.P. For Review of Actions taken by Various National Securities Exchanges and National Market System Plans in Their Role as Registered Securities Information Processors, Exchange Act Rel. No. 84433 (Oct. 16, 2018);
 We recognize that other analyses may also be relevant.
 Rule 611 of Regulation NMS.
 See "Dissent of Commissioners Cynthia A. Glassman and Paul S. Atkins to the Adoption of Regulation NMS" (June 5, 2005), https://www.sec.gov/rules/final/34-51808-dissent.pdf; see also Daniel M. Gallagher "How to Reform Equity Market Structure: Eliminate "Reg NMS" and Build Venture Exchanges," The Heritage Foundation, Prosperity Unleashed: Smarter Financial Regulation (Feb. 23, 2017).
 See Section 916 of the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010.