On October 16, 2018, the five SEC Commissioners unanimously upheld the Securities Industry and Financial Markets Association’s (SIFMA) challenge to fee increases for “depth-of-book” market data filed by Nasdaq and NYSE Arca. Simultaneously the SEC remanded over 400 market data fee and other filings back to the exchanges and the national market system plans for consideration under the standards set out in the SIFMA order and an earlier Bloomberg order. These orders have the potential to rewrite the regulation of market data, and potentially other exchange fees, for the entire securities industry. Sidley represented SIFMA in its successful challenge to the market data fees.
The procedural status of the SIFMA matter is somewhat complex. Under the Exchange Act, fees must be “fair and reasonable” and not “unfairly discriminatory.” Historically, exchanges submitted proposed fee increases to the SEC for approval with limited if any factual support, and the SEC generally routinely approved them without substantial analysis. In NetCoalition v. SEC, 615 F.3d 525 (D.C. Cir. 2010), the court held that competitive forces could be sufficient to meet the “fair and reasonable” standard of the Exchange Act. Nonetheless, in that case, the court found that the SEC had provided insufficient factual support for its conclusion that exchange market data fees were in fact constrained by significant competitive forces.
After substantial procedural wrangling, the SEC selected two market data filings, the NYSE Arca at issue in NetCoalition and a separate Nasdaq filing, for an unusual fact-finding hearing by a commission administrative law judge concerning whether those fees were “fair and reasonable,” including the issue of whether market data fees were constrained by competition. Both filings proposed substantial price increases for depth-of-book market data concerning securities traded on those respective exchanges — data not available from any other source. In the meantime, the SEC held in abeyance more than 400 challenges filed by SIFMA and Bloomberg LP to market data price increases filed by exchanges and the national market system plans (which administer core “top-of-book” market data required to be provided to all investors).
In general, the exchanges made two arguments to the SEC. First, they argued that because the exchanges undeniably compete for order flow, the ability of broker-dealers and other market participants to choose the exchanges or other trading platform to which to route orders constrains any exchange’s ability to increase fees for its own market data. The SEC found that the exchanges’ very limited anecdotal evidence of broker-dealers threatening to divert order flow was unpersuasive. And, in a world in which Regulation NMS requires firms to execute against displayed quotes at every exchange, and obtaining quality executions requires firms to consider depth-of-book interest at all exchanges, the SEC found that the exchanges had not demonstrated that competition for order flow meaningfully constrains prices for market data. Second, the exchanges argued that market participants have the ability to substitute depth-of-book market data from one exchange for that of another, and this process constrains the ability of any single exchange to charge excessive fees. The SEC found that many significant market participants in fact need to buy market data from every exchange, and the data showed that the market for this data is highly inelastic — competition between exchanges does not provide a meaningful check on prices. As a result, the SEC set aside both fee increases, concluding that the exchanges had not met their burdens of showing that the fees met the Exchange Act requirements. Nasdaq has already filed an appeal of the SEC’s decision in the U.S. Court of Appeals for the D.C. Circuit, and NYSE Arca has announced an intent to do so as well.
The SEC did not set aside the remaining 400 challenged market data fee filings. Rather, the SEC remanded them to the exchanges and the national market system plans for reconsideration. The order gave the exchanges and NMS plans six months to “develop or identify fair procedures for assessing the challenged rule changes as potential denials or limitations of access to services.” Those procedures must give affected parties notice and an opportunity to be heard, and allow the exchanges to develop a record and to explain their conclusions based on that record in a written decision sufficient to enable SEC review. Within another six months, the exchanges and NMS plans must apply those new procedures to the remanded fee applications. This SEC remand order contemplates that the exchanges will publish proposed fee filings and that parties potentially affected by fee filings could comment on those proposed changes. How the exchanges will implement this remand order is unclear, and they have sought reconsideration of that order by the SEC.
The SEC’s order in the SIFMA case is relatively narrow. It does not find whether the fees at issue are fair and reasonable; it simply concludes that Nasdaq and NYSEArca did not meet their burden of proof concerning those fees. It does not find that the market for depth-of-book data is uncompetitive; again, it simply finds that the exchanges did not meet their burden of proof concerning the constraining effects of competition. And, the order sets aside only two specific fees: the 2010 ArcaBook Fee Rule and Nasdaq’s 2010 extension of Level 2 fees. Commissioners Hester Peirce and Elad Roisman, while joining in the SEC opinion, issued a concurring statement leaving open an argument in favor of market data fee increases based on client segmentation, and the “platform competition” argument that NYSE Arca advanced. Those Commissioners also expressed a disinclination to regulate market data fees based on the costs actually incurred by the exchanges, which they characterized as requiring the SEC to act as a rate setter.
Nevertheless, the SEC’s SIFMA order appears to raise the bar significantly for what exchanges must show to justify fee increases. On the record before it, the SEC rejected the exchanges’ statistical analyses as showing only correlation, not causation, and for omitting relevant variables from the regression analysis. The SEC rejected reliance on isolated anecdotes of traders’ threats to divert order flow based on the market data fee increases. The SEC closely scrutinized, and rejected, other arguments as insufficiently supported or as based on unreliable evidence.
In addition, exchanges have had their fee structures challenged in contexts other than market data. For example, the SEC rejected the exchanges’ proposed plan for funding the consolidated audit trail.That funding plan would have put almost 88 percent of the costs for the consolidated audit trail system on broker-dealers and alternative trading systems rather than on the exchanges themselves. Second, the SEC has proposed an access fee pilot, which seeks to examine whether exchange maker/taker pricing models undermine the duty of best execution. The exchanges have heatedly opposed the access fee pilot.
More broadly, the SEC (across the broad ideological spectrum of its Commissioners) may be rethinking the role of for-profit exchanges in the regulatory structure. For example, Commissioner Robert Jackson recently called for the SEC to reexamine whether exchanges should have immunity and the scope of that immunity. The Peirce/Roisman concurrence in the SIFMA matter urged the SEC to reexamine the trade-through provisions of Regulation NMS. Before the financial crisis, the SEC had issued a concept release on exchange self-regulation, which pointed out many of the inherent conflicts of interest in having for-profit exchanges conduct self-regulatory functions. SEC Division of Trading and Markets Director Brett Redfearn appears interested in revisiting those issues. The SIFMA order may herald a much broader rethinking of the relationship between exchanges and other market participants.