Skip to main content

Modernizing U.S. Equity Market Structure

Washington D.C.

June 22, 2020

Chairman Clayton:

Good afternoon and welcome to this SEC-sponsored virtual forum to discuss equity market structure.[1]  I want to begin by thanking Deborah Lucas for her kind introduction and Assistant Attorney General Makan Delrahim for giving us his insightful perspective.  Promoting competition is a core objective of U.S. antitrust and securities law, and I very much appreciated hearing his insights.

Today marks the third consecutive year that Director Redfearn and I have addressed equity market structure.  Our remarks have focused on three areas needing close attention—improving the market for thinly traded securities, combatting retail fraud, and addressing concerns about the quality and cost of market data.[2]  Here I am very pleased that we have meaningful progress to report as well as some key initiatives we are working to complete.  I would like to take a moment to commend the Commission’s dedicated staff for continuing to serve the public without missing a beat during these challenging times.

As we did last year, Brett and I will speak in sequence.  I will introduce the issues and provide my thoughts, and then Brett, with perhaps an interruption from me, will provide further details.  Each of us is speaking for ourselves.  My views are my own, Brett’s views are his and, to the extent we discuss potential future action, please know there will be significant discussion, analysis and evolution before any action is taken.

I. Principles Guiding Our Equity Market Structure Agenda

When discussing equity market structure, I have made it a habit to begin by emphasizing the principles that guide our approach.  These principles are clearly articulated in my prior speeches, so I will break that habit and not repeat them here, but they are in my posted remarks.[3]  I am pleased to announce that work pursuant to one key principle—coordinating and communicating with other regulators—is being furthered through a new Memorandum of Understanding signed between the SEC and the DOJ Antitrust Division.  Our agencies each have their respective areas of responsibility—the DOJ is responsible for antitrust policy and enforcement, and the SEC is responsible for securities market function, structure and enforcement—but there are significant commonalities of facts and expertise as both agencies work to promote competitive market conditions.  I believe that close coordination and communication between us—and the experts on our career staff—contribute to a well-functioning regulatory environment.  I know the SEC can benefit from the DOJ’s expertise, and the insightful remarks that Makan delivered earlier serve as proof of that point.

I would also like to reiterate that the power of choice and competition is crucial and formative to our securities markets.  Another of my key principles dictates that access to material information can empower investors and energize the competitive forces that benefit markets broadly.  It is through this lens that I consider our equity market structure and the essential question of whether access to markets and material information about those markets is fair and reasonable.  As I have previously noted, it has long been recognized that market prices can function as “public goods.”[4]  Efficiently providing this function—through a combination of regulation and market forces—can be challenging to achieve in practice, however, particularly in a complex, high-speed environment such as trading.

It is only fair to note that this discussion of improving efficiency and function should be viewed against the backdrop of substantial progress over the past two decades, particularly from the perspective of trading costs.  Today, retail investors pay substantially less for execution than they did ten and twenty years ago and, in our largest stocks and related derivatives, liquidity is strong.  We also should remember that liquidity is not free—market makers and other professional traders participate actively because they expect to turn a profit.  Here, I would be remiss if I did not mention the recent court case vacating the Commission’s transaction fee pilot.  While I think that the Commission should continue to focus on and pursue data-driven analysis, rulemaking and policy, I accept the decision of the D.C. Circuit and appreciate the guidance it provides.  I expect the Commission will move forward with its efforts to improve and modernize our National Market System following that guidance.  To me, the decision served to emphasize our need to have real data from exchanges, ATSs, and other market participants to facilitate oversight and analysis of new and existing rules.  The court has said that the Commission cannot set up this kind of controlled environment, and so I expect we will continue to work in the real environment to make sure we have the data and other information we need.[5] 

II. Modernization of Equity Market Structure

The initiatives we will discuss today are intended generally to improve our ever-changing securities markets.  We began this journey in 2018 by identifying three market structure areas in need of modernization.  For each of these areas, we have followed the same transparent and rigorous path forward:  the staff held roundtables that sought out a wide range of viewpoints and then developed specific initiatives to advance for public comment.  I am pleased to say that the public comments we have received reflect a breadth and depth of expertise and insight on what can be highly complex and technical market structure issues. 

Improving the Market for Thinly Traded Securities

One area we have targeted for progress is improving the quality of our market for thinly traded securities.  Today, Regulation NMS mandates a single market structure and regulatory framework for all exchange-listed stocks, regardless of their different characteristics.  At the 2018 Roundtable on Market Structure for Thinly-Traded Securities, several participants were critical of this one-sized-fits-all approach and highlighted the particular challenges facing companies and investors in this segment of the market.[6]

Following the Roundtable, in October 2019, the Commission published its Statement on Market Structure Innovation for Thinly Traded Securities inviting market participants to submit innovative proposals designed to improve the secondary market for thinly traded securities, including, in connection with such proposals, requests to suspend or terminate unlisted trading privileges, known as UTP.[7]

Before I turn to Brett to discuss some of the details of this initiative, I want to make a few points.  First, the SEC’s effort to improve secondary market trading for thinly traded securities is but one of several initiatives that the SEC has pursued to advance the interests of small and mid-size companies.[8]  And second, there is no magic solution that will suddenly produce deep pools of liquidity for thinly traded securities.  But the fact that the task may be difficult is no reason not to take it head on.  We must think creatively about how best to proceed and without letting the perfect be the enemy of the significantly better.

Director Redfearn:[9]

At the same time that the Commission issued its Statement, the Division of Trading and Markets published a Background Paper on the Market Structure for Thinly Traded Securities that presents some of the characteristics and trading challenges of this segment of the market.[10]  Companies with stocks in this segment of the market face some difficult challenges not faced by companies with actively traded stocks.  The Staff Background Paper references economic research indicating that improving secondary market liquidity can have real benefits for companies.  I believe there are serious questions, however, about whether the current market structure that works relatively well for very active stocks is optimal for thinly traded securities.

The Commission’s Statement mentions a number of potential innovations that may be worth considering.  These include providing incentives to market makers to assume heightened obligations with regard to thinly traded securities, implementing periodic intraday auctions as a means to concentrate liquidity, and introducing non-automated markets to facilitate trade negotiation and incentivize market maker participation. 

I am pleased that a range of commenters have responded to the Statement and submitted views and potential approaches, including an application by an exchange for an exemption from UTP, to improve the market structure for thinly traded securities that the staff is studying closely.[11] 

Going forward, proposals should fully lay out the elements of the proposal and the rationale for whatever relief is requested from current rules.  To the extent proposals involve proposed rule changes, these would provide an opportunity for the public to comment on the specifics of an exchange’s proposed market structure innovations, and would not become effective unless the Commission approved the proposed rule change. 

Chairman Clayton:

Combatting Retail Fraud

We have taken a comprehensive approach to combatting retail fraud and protecting the interests of our long-term Main Street investors.  Among other things, we have focused on empowering investors by providing them with the information and tools they need to identify and avoid fraudsters—in other words, to help prevent harm rather than seeking to remedy harm after-the-fact.

At the 2018 Roundtable on Combatting Retail Fraud, one of the main topics of discussion concerned investors’ lack of sufficient information about companies with securities that are not listed on a national securities exchange, or OTC securities.[12]  Over decades, many of these types of securities have served as vehicles for fraud and manipulation, and just over the last several years, the Commission has brought hundreds of enforcement actions involving OTC securities or their issuers.[13]  This has gone on too long.

What adds to my concern is that securities that trade in the OTC market are primarily owned by and marketed to retail investors.  Just as material information and transparency serve as the foundation for our federal securities law framework, the information that is available to these investors trading OTC securities is a vital element of our regulatory scheme to protect retail investors.  Under Rule 15c2-11, broker-dealers serve as key “gatekeepers” in helping to prevent fraud and manipulation in OTC securities.  The Rule sets out requirements that broker-dealers must meet before they can publish quotations in an OTC security.  These broker-dealer quotations are a primary mechanism for facilitating the secondary market trading of OTC securities.

At the Roundtable, panelists were concerned that Rule 15c2-11 was outdated and needed to be modernized in several respects.[14]  Well, the Rule was last amended in 1991.  Clearly, much has changed in the financial markets since then, and the Rule needed to be closely reviewed.  Today, among other things, the internet provides an incredibly cost-effective and efficient mechanism to collect and make information publicly available.

One of the major concerns about the Rule is its “piggyback” exception.  Under the piggyback exception, the current Rule allows quoting in an OTC security to continue in perpetuity, even when there is no or limited current, publicly available information about the issuer and even when the issuer no longer functions or even exists.  Experience tells us, unfortunately, that these information deficiencies can be fertile ground for fraud.

I am pleased that last September, the Commission proposed amendments to Rule 15c2-11.[15]  The amendments are designed to increase the availability of issuer information and modernize the rules governing quotations for OTC securities.  I agree that sunlight is the best disinfectant, and in particular, I believe that we owe it to our Main Street investors to ensure that a minimum level of current information is publicly available if securities are to be quoted by brokers.  Investors can and should be able to continue to trade OTC securities, but they should be able to do so with reasonably current—and not stale—information.

Brett will discuss some of the other details of the Commission’s proposal, but I first want to express my appreciation for the more than 150 comments submitted by the public.[16] 

Director Redfearn:

As Chairman Clayton covered in his remarks, a primary element of the 15c2-11 proposal is to enhance transparency for OTC securities by requiring that information about the issuer and its security be current and publicly available.  Given the tremendous advances in information technology since 1991, the proposal is designed to modernize the Rule by leveraging these advances to minimize the cost of compliance beyond what was previously possible.

Another key element of the proposed amendments to the rule is designed to reduce regulatory burdens on broker-dealers quoting those types of OTC securities that may be less susceptible to fraud and manipulation.  These include, for example, actively traded securities of well-capitalized issuers and securities for which a broker-dealer publishing the quotation was a named underwriter.  For these securities, the proposal would eliminate the review requirement and provide an on-ramp to a broker-dealer quoted market.  The proposed rule would also increase efficiency by allowing broker-dealers to rely on centralized determinations regarding whether an exception is available.

An overriding objective of the Rule 15c2-11 proposal is to enhance transparency and investor protection, while preserving the efficient access of investors to valuable investment opportunities.  The commenters on the proposal provided helpful views on how to achieve the dual objectives of investor protection and efficient access.  The staff is considering these views as we develop our recommendation to the Commission.

Chairman Clayton:

Modernizing NMS Market Data and Access

The third roundtable initiative relates to market data and access.  Market data is the fundamental source of transparency and price discovery for the secondary equity markets.  And by market data, I mean real-time information concerning the prices at which securities can be traded and the prices of trades that already have been executed.  Collecting, consolidating, and disseminating this data have formed the heart of the National Market System ever since Congress mandated its creation in 1975.  In 2005, in Regulation NMS, the Commission emphasized that NMS market data enabled “investors of all types—large and small—[to] have access to a comprehensive, accurate, and reliable source of information for the prices of any NMS stock at any time during the trading day.”[17]

Currently, we have (1) NMS market data that is disseminated by an exclusive processor, known as the SIP, which the exchanges govern jointly pursuant to three separate NMS plans; and (2) an array of proprietary data products that the exchanges sell to various market participants.  NMS market data, generally speaking, informs the retail segment of the market as well as professional traders looking at screens and making investment decisions at speeds handicapped by the response times of the human brain.  The proprietary data products generally are faster and provide richer and more detailed trading data than NMS market data.  These products are also essential fuel for the trading systems used by electronic market makers and other firms with highly sophisticated technologies and trading strategies that find opportunities in microseconds—and increasingly nanoseconds. 

Many panelists and commenters at the 2018 Roundtable on Market Data and Market Access expressed a strong belief that NMS market data was no longer adequate to meet the needs of many investors, both retail and institutional.[18]  Their criticisms were wide-ranging and included concerns about the declining relative utility of the content of NMS market data as trading practices and regulations changed, as well as the measurably slower timeliness as technology evolved and trading speed increased.  Various market participants stated that, given what they viewed as the shortcomings of NMS market data, they felt compelled to purchase proprietary depth of book data products.[19]  Panelists and commenters also were concerned that deficiencies in the governance of NMS market data had contributed to the SIPs’ comparatively slower evolution.

The SEC has emphasized that one of its “most important responsibilities is to preserve the integrity and affordability of” NMS market data.[20]  Given the fundamental concerns raised at the Roundtable by a broad spectrum of investors and market participants, I asked the staff to develop recommendations for Commission initiatives to address these concerns.

I am pleased that the Commission was able to move forward with three initiatives over the last year.  These initiatives addressed (1) the process for review of NMS market data fee changes, (2) governance of the NMS market data plans, and (3) infrastructure for NMS market data.  Once again, I will turn it over to Brett to discuss the details of these initiatives after emphasizing a few points.

First, Main Street investors require NMS market data, at a minimum, to participate in the U.S. equity markets.  The CTA plan website, for example, reveals that 3.48 million non-professionals were monthly subscribers to data for NYSE-listed securities in the fourth quarter of 2019.[21]  Many millions more use NMS market data on a “per-quote” basis.[22]  These users, whether monthly or per-quote, generally are retail investors, obtaining data by requesting quotes from their brokers.  

Second, I am pleased that the infrastructure proposal would introduce, for the first time, competitive forces into the model for processing and distributing NMS market data.  Under the proposal, the current exclusive SIP model developed in the 1970s would be replaced by a model that (1) accommodates multiple competing consolidators, and (2) would allow firms to process, or “self-aggregate,” NMS market data feeds, in a way that is similar and consistent with the way in which firms self-aggregate proprietary data feeds today.  Significantly, the proposal would also create round lot tiers where lot size better reflects the prices of securities and the notional value of the posted quote.  In today’s markets, the average consumer is often not quoting or trading in 100-share round lots, and sometimes, in these circumstances, price improvement may be illusory.

And finally, while there are undoubtedly many issues to be addressed in terms of whether initiatives as proposed should be modified or improved, I do not believe that the status quo is acceptable.  I anticipate that considering each of the three market data initiatives will be a high priority item on the SEC’s equity market agenda for the remainder of the year.

Director Redfearn:

Rescinding the Effective-on-Filing Procedure for NMS Plan Fee Changes

The first market data proposal was published last October and addresses the procedure for review and approval of NMS plan fee changes.[23]  These fees currently are charged by the NMS plans for market data in NMS stocks and listed options, and they are substantial—the fees exceeded $500 million in 2017 and the NMS plans charge for their use by a broad spectrum of market participants.[24]  These include millions of retail investors, institutional investors, professional traders, broker-dealers, and vendors.

Currently, however, Rule 608 of Regulation NMS provides an exception to the normal procedure for review and approval of NMS plan amendments.  The exception allows fee changes to be immediately effective upon filing with the Commission, prior to an opportunity for public vetting and without Commission review or approval.  Under the proposal, the fee exception would be rescinded.  Public notice and comment, in particular, is an integral aspect of the Commission’s process for assessing important policy questions across many contexts.  I anticipate the Commission soon will consider a staff recommendation on rescinding this exception.

Improving Governance of the NMS Plans for Equity Data

The second market data proposal was first published in January and addressed the governance structure of the three separate NMS plans for equity market data.  The Commission proposed requiring the self-regulatory organizations, or SROs, that are participants in the three NMS plans to propose a new single plan to govern the arrangements for public dissemination of NMS market data.[25]  After reviewing many thoughtful comments on the proposed order, the Commission issued a Final Order last month.[26]

The Final Order begins by describing some of the significant content and latency deficiencies of NMS market data that Chairman Clayton noted earlier.  One fundamental problem stems from the inherent conflict of interest that exists for exchanges that offer competing proprietary data products at the same time as they are responsible for the governance and operations of NMS market data.  Another is the limitation on voting participation in the NMS plans to SROs exclusively, which excludes all other stakeholders in NMS market data.

To address these concerns, the Final Order requires the SROs to submit a proposed new NMS plan that will make adjustments to the voting rights for exchange groups with multiple SROs and expand voting participation to key stakeholders with a diversity of views.  The new plan also will require provisions to help manage the conflicts of interest of exchanges that offer proprietary data products that compete with NMS market data.

The Commission’s order directed the SROs to submit a new plan within 90 days, which will be noticed for public comment before the Commission takes action. Until a new plan has been approved by the Commission, the current equity data plans will continue in effect.  Both NASDAQ and NYSE have filed petitions for review of the Final Order in the D.C. Circuit. 

Upgrading the Infrastructure of NMS Market Data and Access

The Commission’s third market data initiative proposes meaningful changes to modernize the infrastructure for collecting, consolidating, and disseminating NMS market data.[27]  The proposal is designed to update the content of the information with respect to quotations for and transactions in NMS stocks and to introduce a decentralized consolidation model for the collection, consolidation, and dissemination functions currently performed by the exclusive SIP.

The proposal would greatly expand the content of NMS market data in three ways:  (1) lowering the round lot size to improve pre-trade transparency for many higher-priced securities, (2) including depth-of-book price levels beyond the best bid and offer, and (3) including information that facilitates participation in exchange auctions.

I want to focus first on the proposal to lower the round lot size for many higher-priced securities.  Currently, the round lot size for nearly all NMS stocks is 100 shares, regardless of its price.  This means, for example, that the best available quote for an NMS stock with a price of $10 must have a quoted dollar value of at least $1,000, while the best available quote for an NMS stock with a price of $500 must have a quoted dollar value of at least $50,000.  Further, the quoted spreads currently visible in NMS market data for higher-priced stocks do not include significant quotation information of smaller lot sizes that is visible only in certain proprietary data feeds.  NMS quotations are therefore less representative and significantly wider than they would be if smaller dollar-sized quotations were included.

As noted, exchange proprietary data feeds typically include all odd-lot quotes and thereby provide better and more thorough information than is available in the NMS market data feeds.  For example, the proposal includes empirical analysis showing that, for stocks priced over $1000, the proposal to lower the round lot size could improve quoted spreads in NMS market data 92 percent of the time and narrow the width of such spreads by more than half.[28]  Today, the many retail investors that use NMS market data cannot see the better prices available to those who can pay the much higher fees for exchange proprietary data.  Furthermore, execution quality statistics, as provided under Rule 605, also only use the wider 100 share round lot quotations, which can affect the view of price improvement.  Enhancing the quote information in NMS market data would be quite useful to retail investors in assessing the quality of executions obtained by their broker-dealers.

To this end, the proposal would establish five tiers of round-lot quotes.  NMS stocks with prices of $50 or less would retain the 100-share round lot, while NMS stocks with prices higher than $50 would have four progressively lower tiers of round-lot sizes to maintain a relatively consistent dollar size for the inclusion of quotes in NMS market data, regardless of stock price.  A key objective in formulating these tiers was achieving the right balance of, on the one hand, capturing additional liquidity to improve the usefulness of quote information in NMS market data and, on the other hand, avoiding unnecessary complexity with too many tiers. 

The importance of achieving the right balance must not be overlooked, particularly from the standpoint of retail investors, who may not know that they may be receiving prices inferior to odd-lot quotes.  An example may help illustrate this point.  Today, the best-priced offer for a higher-priced stock in NMS market data may be 100 shares for $400.50, while the exchange proprietary data feeds may have a 20-share offer at $400.45.  If a retail investor places a 20-share buy order and it is executed at $400.48, the executing venue is entitled to report that the retail investor received two cents for price improvement.  This is true even though an odd-lot quote on the exchange proprietary data feeds was readily available at a price that was three cents better than the “price-improved” execution supposedly provided to the retail investor.  Because the NMS market data does not show the true best-priced quote for her order, the retail investor is none the wiser that in fact she received a price that was three cents inferior than what was readily available in the market.  In striking the right balance between data usefulness and complexity, we must not shortchange the information needs of retail investors.

Another aspect of the proposal is that if adopted as proposed it would not expand the scope of the trade-through rule, Rule 611 of Regulation NMS, to the new round-lot quotes.  A “trade-through” is the execution of a trade at a price that is outside of the best displayed quoted price, which is protected by regulation.  This is another area where a key objective is to strike the right balance.  We are considering whether the highly prescriptive trade-through rule is necessary for these smaller quotations, or whether an approach that relies on competitive forces and the duty of best execution better strikes the right balance of protecting investors, particularly retail investors, without introducing unnecessary complexity and rigidity.  As expected, many commenters have focused on the issue of order protection for the new round lots, and we are carefully considering all comments.

Beyond expanding and improving the content of NMS market data, a second major objective of the infrastructure proposal is to reduce its latency when compared to the exchange proprietary data feeds.  The current operating model mandates an exclusive processor for NMS stocks in a centralized consolidation model, whereby data from multiple exchanges in different locations is sent to an exclusive processor in one location, aggregated, and then disseminated back to markets and market participants in multiple, often distant locations.  This centralized model adds a significant amount of travel time between locations, which we refer to as geographic latency.  The proposal would reduce this latency by allowing for a decentralized distribution model and introducing competition into the model for aggregating and disseminating NMS market data.  The proposed model allows for exchange data to be sent directly to competing consolidators or self-aggregating firms in multiple locations to avoid this unnecessary added latency.  For years, the private data market has utilized this form of direct data distribution due to its material latency benefits.

A potential objection to a competing consolidator model is that it would mean losing the “single NBBO” under the current exclusive processor model.  “NBBO” stands for “national best bid and offer.”  But, as the Commission explained in its proposal, the idea that there is only one NBBO does not reflect today’s reality.  Many market participants calculate their own NBBO from proprietary data feeds and these vary ever so slightly depending upon, for example, the exact location of the participant. The concern about the loss of a single NBBO that is exactly the same regardless of location simply does not acknowledge common practices and physical realities in our markets today.  On these practices, I will share a few key points:

  1. Because many trading firms do not consider NMS market data to be competitive, those firms often purchase proprietary data directly from exchanges.  These firms also aggregate that data into an NBBO that is used for both trading and regulatory purposes.  And, since these firms are often located in different data centers, they are also subject to the laws of physics. 
  2. On physics, there are unavoidable limitations due to the speed of light—and the speed of fiber optic cable.  This means that, as long as users are not all in the exact same location—which they are not—information cannot reach all users at exactly the same time.  As a result, the NBBO at one location will vary slightly from the NBBO at another location. 

Today, hundreds of firms are located in different locations.  Some buy NMS market data, some aggregate proprietary data.  Some get their data over fiber.  Some get their data over microwave towers.  Anyhow, you get the point.  There is no one NBBO in a world where markets are quoting and trading in nanoseconds.

The NMS rules recognize, as they must, this reality and allow market participants to trade based on information as it arrives.  In the 2005 adopting release for Regulation NMS, for example, the Commission explicitly stated that a trading center’s compliance with Rule 611 is based on when information arrives at that trading center.[29]

As noted, the proposed competing consolidator model would not change the fact that there are many NBBOs today.  It would, however, introduce competitive forces to help ensure that NBBOs reflect a wider range of orders and are available to a broader range of market participants at lower latency than they are under today’s monopoly model.  This rule would better enable the vast array of competitive technologies that have been deployed in the proprietary data market also to be deployed in a competitive fashion in the market for NMS market data.

Chairman Clayton:

III. Helping to Ensure SRO Market Data and Connectivity Fees Comply with Exchange Act Standards

The last topic we will discuss today is the SEC’s role in ensuring that exchange fees for market data and connectivity comply with the Exchange Act.  Under Section 19(b) of the Exchange Act, Congress charged the SEC with the responsibility to determine whether the fees charged by exchanges comply with statutory standards[30] including, as Brett will describe in greater detail, that fees be fair and reasonable and equitably allocated across users of an exchange’s facilities. 

As you may know, a recent judicial decision addressed the Commission’s review of exchange fees and ended what had been a 14-year saga that has played out in the Commission and the courts.  That is too long.  I will try to give you the abridged version.  It all began in 2006, when an exchange filed a rule change to impose a fee for its depth-of-book product.  The Commission approved the fee change, applying a “market-based approach” to determine whether the fee was fair and reasonable under the Exchange Act.  In 2010, market participants challenged the Commission’s approval, and the D.C. Circuit vacated the Commission’s order, finding that the administrative record did not sufficiently support the conclusion that competition would appropriately constrain the exchange’s depth-of-book fees.  In late 2018, after consolidating two exchanges’ depth-of-book fee filings in response to challenges by market participants, the Commission found that the two exchanges had failed to demonstrate that competition appropriately constrained the exchanges’ pricing of their fees.  Separately, between the 2010 D.C. Circuit decision and the 2018 Commission order, Congress enacted a new law under the Dodd-Frank Act that made exchanges’ fee changes immediately effective upon filing.  Following this procedural shift, market participants challenged an additional 400 fee changes.  This month, the D.C. Circuit vacated the Commission’s 2018 order. 

But this time, the Court’s decision narrowly focused on a procedural issue.[31]  The Court held that generally applicable fees for market data charged by the exchanges—these are fees that are applicable to all market participants—could not be challenged by market participants in a particular type of administrative proceeding, known as a denial of access proceeding. 

The court did not address a key substantive issue in the case, namely, the Commission’s finding that two exchanges had failed to meet their burden of demonstrating that their fees for proprietary market data were constrained by competitive forces and therefore met Exchange Act standards. 

So where does that leave us?  One thing is crystal clear.  In the absence of evidence demonstrating that competitive forces actually constrain the pricing of market data, the Commission has the obligation, under the Exchange Act, to suspend exchanges’ fee filings unless it is established that the fee is reasonable on another basis, such as a reasonable cost basis.  In other words, the exchange has the burden of demonstrating these competitive forces or an alternative basis for finding the fee fair and reasonable.

Many panelists and commenters at our Market Data Roundtable noted that proprietary exchange fees had risen significantly, did not believe that the exchange fees were constrained by competitive forces, and questioned whether the fees were set at levels consistent with Exchange Act requirements to be reasonable.[32]  The DOJ’s Antitrust Division, led by Mr. Delrahim, submitted a thoughtful comment letter on the infrastructure proposal.[33]

Clearly, given the diversity of views, the Commission has a compelling regulatory responsibility to analyze concerns about the fairness and reasonableness of exchange fees for proprietary data.  I believe there is a clear need for prompt diligent effort, resourcefulness, and collaboration regarding ways for the Commission to fulfill this statutory responsibility in the most efficient and effective manner.

For example, the exchanges have offered various rationales and economic theories for their assertions that competitive forces discipline their proprietary data fees.  As these fees continue to come under review in the various procedural contexts, I expect the Commission will be open to any reasonable framework for demonstrating sufficient competition.  In addition, in a complex market, it is unlikely that any one metric would be dispositive.  I expect that the exchanges will be forthcoming in presenting meaningful data and other factual support to meet their burden of demonstrating that the fees are consistent with the Exchange Act.  I have also asked the staff to focus on these issues and report back to the Commission on potential approaches that could be adopted to fulfill its statutory responsibilities for exchange fees.  There are many different approaches that economists have taken to show competition, here at the SEC and at the DOJ Antitrust Division and elsewhere.  I look forward to using the new MOU with the DOJ Antitrust Division to work in coordination where necessary to benefit from the views of the DOJ staff on certain competitive theories to assess whether they are consistent with Exchange Act obligations. 

Director Redfearn:

As Chairman Clayton noted, the Exchange Act prescribes a series of substantive standards for SRO fees.  These standards primarily are set forth in Section 6 for exchanges, in Section 15A for FINRA, and in Section 11A for market data fees.  For simplicity’s sake, I will focus today on the fees of exchanges and, in particular, on Section 6(b)(4), which requires that an exchange’s rules “provide for the equitable allocation of reasonable dues, fees, and other charges among its members, issuers, and other persons using its facilities.”  Three key elements of this standard are that (1) an exchange’s fees be set at a reasonable level, (2) the fees be equitably allocated across users of the exchange’s facilities, and (3) the fees impose no undue burden on competition.

Division staff has been highly focused on the Commission’s regulatory responsibilities for helping to ensure that exchange fees for market data and connectivity meet the relevant statutory standards.  Pursuant to its delegated authority to act on behalf of the Commission, the staff has suspended certain proposed fee changes pursuant to Section 19(b) and instituted proceedings to assess whether exchanges have met their burden of demonstrating that proposals meet the Exchange Act standards for fees.  And courts have instructed us that “unquestioning reliance” on the representations of an SRO is not a sufficient basis for the Commission to make this finding.[34]

The Staff posted guidance in May 2019 to assist SROs in preparing fee filings that meet their burden of demonstrating compliance with Exchange Act standards.[35]  This guidance of course only reflects staff views and is not a rule, regulation, or statement of the Commission.

The Staff Fee Guidance notes that the Commission traditionally has taken a market-based approach when assessing the reasonableness of exchange fees.  Under this approach, the Commission should begin by examining whether an exchange was subject to significant competitive forces in setting the terms of its proposal, including the level of fees.[36]   

Assessing whether competition at the platform level constrains an exchange’s overall fees is a factual issue for which financial information concerning the exchange’s revenues, costs, and margins would be highly relevant.  Currently, the best publicly available financial information for exchanges is provided by their annual updates to Form 1, which is the form for registration of a national securities exchange.[37]  Although the usefulness of Form 1 financial information is limited in many respects, it is sufficient to demonstrate a substantial shift in the source of exchange revenues during the period from 2013 to 2018.  For seven of the largest volume equity exchanges that operated throughout that period, their revenues from market data and connectivity rose at least 65 percent,[38] while the net revenues from transaction services rose by at most 13 percent.[39]  During this same period, exchange operating expenses declined by 4 percent, while exchange operating income increased by 106 percent.

In this context, these changes raise questions about whether the current level of various exchange fees for market data and connectivity are reasonable, as well as whether these fees are equitably allocated across users of exchange facilities.  Moreover, the fundamental question as to whether these fees are subject to competitive forces remains an important and concerning question.

In light of the continuing concerns over whether exchanges have demonstrated the fairness, reasonableness and equitable allocation of fees for market data and connectivity, Division staff is considering a variety of approaches for assessing the level and allocation of SRO fees.  We have been and will continue to gather and evaluate relevant data, including potentially data related to profitability, return on assets, or other metrics, or a combination of metrics, as appropriate.  And we will continue to assess theories that potentially bear on these issues.  In the meantime, through the rulemaking proposals previously discussed, we are attempting to introduce a world of data that is sufficient in content, competitive in speed, and subject to market forces, with data fees that are subject to notice and comment and Commission approval.

Chairman Clayton:

IV. Conclusion

I will conclude these market structure remarks in the same way as the last two years, and that is by asking for the continued assistance of the public, including investors, broker-dealers, exchanges, vendors, and all other stakeholders.  As Brett and I have discussed, the SEC has sought comment on a series of significant market structure initiatives, and the public comment file plays an essential role in the rulemaking process.  I look forward to your continued assistance as we move forward on equity market structure.

Thank you.


[1] My words are my own and do not necessarily reflect the views of my fellow Commissioners or the SEC staff.

[2] See Chairman Jay Clayton and Director Brett Redfearn, Equity Market Structure 2019: Looking Back & Moving Forward (Mar. 8, 2019) (“2019 Remarks”), available at; Chairman Jay Clayton, Remarks at the Equity Market Structure Symposium Sponsored by the University of Chicago and the STA Foundation (Apr. 10, 2018), available at; Director Brett Redfearn, Remarks at the Equity Market Structure Symposium Sponsored by the University of Chicago and the STA Foundation (Apr. 10, 2018), available at

[3] Those principles are: (i) fidelity to the SEC’s mission to protect investors, to maintain fair, orderly, and efficient markets, and to facilitate capital formation; (ii) focusing on the long-term interests of Main Street investors; (iii) facilitating public transparency that can energize competitive forces to benefit investors; (iv) retrospectively reviewing Commission rules to assess whether they are functioning as intended, particularly as technology spurs new market mechanisms and trading practices; and (v) coordinating and communicating with other regulators and the stakeholders in the markets that we oversee.  2019 Remarks, supra note 2.

[4] See Chairman Jay Clayton, Remarks to the Economic Club of New York (Sept. 9, 2019), available at also, e.g., Paul A. Samuelson, “The Pure Theory of Public Expenditure,” The Review of Economics and Statistics, Vol. 36, No. 4 (Nov. 1954), pp. 387-389; Paul A. Samuelson, “Diagrammatic Exposition of a Theory of Public Expenditure,” The Review of Economics and Statistics, Vol. 37, No. 4, (Nov. 1955), pp. 350-356.

[5] During oral argument, counsel for Nasdaq recognized the Commission’s ability to request data from certain market participants.  Oral Argument, New York Stock Exchange LLP, et al v. SEC, 19-1042 (D.C. Cir. Oct. 11, 2019), available at$file/19-1042.mp3

[6] Panelist views at the Roundtable are summarized in the 2019 Remarks, supra note 2.  The transcript for the Roundtable on Market Structure for Thinly-Traded Securities is available at

[7] Commission Statement on Market Structure Innovation for Thinly Traded Securities, Securities Exchange Act Release No. 87327 (Oct. 17, 2019), 84 FR 56956 (Oct. 24, 2019).

[8] These other efforts have included actions under the JOBS Act to promote emerging growth companies, such as by expanding their opportunities to enter the public capital markets.  See, e.g., Chairman Jay Clayton, Modernizing our Regulatory Framework: Focus on Authority, Expertise and Long-Term Investor Interests (Nov. 14, 2019), available at  Small and mid-size companies have been and will continue to be a priority at the SEC.

[9] The Securities and Exchange Commission disclaims responsibility for any private publication or statement of any SEC employee or Commissioner. This speech expresses the author’s views and does not necessarily reflect those of the Commission, the Commissioners, or other members of the staff.

[10] Division of Trading and Markets, Background Paper on the Market Structure for Thinly Traded Securities, available at

[11] Comments on the Commission’s Statement are available at

[12] Panelist views at the Roundtable were summarized in our 2019 Remarks, supra note 2.  The transcript for the Roundtable on Regulatory Approaches to Combatting Retail Fraud is available at

[13] See Publication or Submission of Quotations Without Specified Information, Securities Exchange Act Release No. 87115 (Sept. 25, 2019), 84 FR 58206, 58210 (Oct. 30, 2019) (“Rule 15c2-11 Proposing Release”).

[14] See supra note 12.

[15] Rule 15c2-11 Proposing Release, supra note 13.

[16] We are reviewing these comments closely as we move forward with this important initiative to promote the interests of retail investors.

[17] Regulation NMS, Securities Exchange Act Release No. 71808 (June 9, 2005), 70 FR 37496, 37557 (June 29, 2005) (“Regulation NMS Release”).

[18] The views of panelists at the Roundtable are summarized in our 2019 Remarks, supra note 2.  The full set of materials for the Roundtable on Market Data and Market Access, including agenda, panelists, a transcript of the panel discussion, and comments from the public, are available at

[19] See, e.g., Transcript of Day One of Roundtable on Market Data and Market Access (Oct. 25, 2018) (“Day One Transcript”), available at, at 65-66 (Mehmet Kinak, T. Rowe Price) (“But as far as brokers having a choice of whether or not they can use the SIP or direct feeds, that doesn’t exist. There is no choice there. If a broker is routing using SIP data, they are not routing my flow. . . . If I’m slower than the other person, I lose. That’s it. That’s the fraction of time we’re talking about. So when someone says, hey, from a commercial enterprise, it makes sense for you to use a faster system over a slower system -- no. This is a best execution obligation. We are obligated to try and produce best execution on every single order that we have. If our brokers are not aligned in that manner to use the most direct, the fastest, the most robust feeds they can get their hands on, then we will trade with someone else.”); id. at 198-199 (Joseph Wald, Clearpool Group) (“Clearpool and other broker-dealers are compelled to purchase exchanges’ proprietary data feeds, both to provide competitive execution services to our clients and to meet our best execution obligations due to the content of the information contained in the proprietary data fees as well as the latency differences between them, which are major and important considerations for brokers.”).

[20] Regulation NMS Release, 70 FR at 37503.

[21] Usage statistics for CTA plan data are available at

[22] In the fourth quarter of 2019, per-quote data usage for NYSE-listed securities was 692 million quotes.  This includes non-professional users that are not classified as monthly subscribers and professional users that choose to report their usage on a per-quote basis.  Id.  We must ensure that NMS market data is sufficient to meet the needs of investors and other market participants as markets evolve and that it is available on terms that facilitate its widespread availability.  This is a primary objective of all three of our initiatives—helping to ensure fair and reasonable fees, expanding governance to include non-SROs, and enhancing infrastructure to improve data quality.

[23] Rescission of Effective-Upon-Filing Procedure for NMS Plan Fee Amendments, Securities Exchange Act Release No. 87193 (Oct. 1, 2019), 84 FR 54794 (Oct. 11, 2019).

[24] See id.

[25] Notice of Proposed Order Directing the Exchanges and the Financial Industry Regulatory Authority to Submit a New National Market System Plan Regarding Consolidated Equity Market Data, Securities Exchange Act Release No. 87906 (Jan. 8, 2020), available at

[26] Order Directing the Exchanges and the Financial Industry Regulatory Authority to Submit a New National Market System Plan Regarding Consolidated Equity Market Data, Securities Exchange Act Release No. 88827 (May 6, 2020), available at

[27] Market Data Infrastructure, Securities Exchange Act Release No. 88216 (Feb. 14, 2020), 85 FR 16726 (Mar. 24, 2020) (“Infrastructure Release”).

[28] See id. at 16823-16824.

[29] Regulation NMS Release, 70 FR at 37523 n. 215 (a trading center’s compliance with Rule 611 “will be assessed based on the times that orders and quotations are received, and trades are executed, at that trading center”).

[30] Under Section 19(b) of the Exchange Act, the Commission cannot approve an SRO’s proposed rule change unless it finds that the proposal is consistent with the Exchange Act.  Section 19(b)(3)(A) provides that a proposed rule change takes effect upon filing with the Commission if the SRO designates the proposal as establishing or changing a fee, but Section 19(b)(3)(C) also provides that the new or changed fee may be enforced by the SRO to the extent it is not inconsistent with the Exchange Act, the rules thereunder, and other applicable law.  Section 19(b)(3)(C) further provides that, within 60 days of filing, the Commission may summarily suspend an immediately effective fee change if it appears to the Commission that such action is necessary or appropriate in the public interest, for the protection of investors, or otherwise in furtherance of the purposes of the Exchange Act. 

[31] See The Nasdaq Stock Market v. SEC, No. 18-1292 (D.C. Cir. June 5, 2020), available at$file/18-1292-1845826.pdf.

[32] See, e.g., Day One Transcript, supra note 19, at 48 (Mehmet Kinak, T. Rowe Price) (“While trading venues, exchanges and so-called dark pools clearly compete for order flow from broker-dealers and this competition has been the primary driver of reduced costs for investors, it is fairly clear that this competition has not constrained the cost of exchanges’ critical market data services.”); id. at 198 (Joseph Wald, Clearpool Group) (“It is very difficult for consumers of market data disseminated by exchanges to understand the reasonableness of pricing due to the lack of information provided by the exchanges around market data and other fees.”); SIFMA Letter on Roundtable on Market Data and Market Access (received on Oct. 24, 2018) (“SIFMA members continue to purchase both SIP and proprietary data despite the growing fees. In particular, the presentation shows that the NYSE has used various types of fee changes to increase the surveyed members’ cost for proprietary market data products by over 1,100% over the last eight years.”), available at

[33] The letter notes that certain information is only accessible via proprietary data feeds, which supply information exclusive to the exchange where the data originates, and that the proposal indicates that increases in proprietary data fees have outpaced increases in NMS market data fees.  The DOJ states that these characteristics “would tend to indicate that Prop Data products lack substitutes, which would in turn enable the exchanges to exercise market power in determining their pricing of these products because they are the only data provider in their own markets.”  DOJ, Comments of the United States Department of Justice (May 26, 2020), available at

[34] See Susquehanna Int’l Grp., LLC v. SEC, 866 F.3d 442 (D.C. Cir. 2017).

[35] Division of Trading and Markets, Staff Guidance on SRO Rule Filings Relating to Fees (May 21, 2019) (“Staff Fee Guidance”), available at

[36] Both the courts and the Commission have recognized that an exchange’s costs may be relevant in determining the reasonableness of fees, even when determining whether competitive forces constrain the fees at issue.  See id. at n. 26.

[37] Form 1 financial information is discussed in the Infrastructure Release, supra note 27, 85 FR at 16817.

[38] The exact amount of exchange revenues derived from market data and connectivity cannot be calculated with Form 1 financial information.

[39] Net revenues from transaction services are calculated by deducting the direct cost of generating such revenues (such as liquidity rebates, brokerage and clearing fees, and Section 31 fees, which are presented separately on exchange financial statements) from total transaction revenues.

Return to Top