I would also like thank the University of Chicago, Jim Toes and the STA Foundation for the invitation to speak at this timely symposium on equity market structure.   

I want to follow up on the remarks of Chairman Clayton on some of the key equity market structure initiatives that are underway and contemplated for the future.  In particular, I would like to briefly elaborate on both current and planned pilots as well as the roundtables on equity market structure that the Division of Trading and Markets anticipates holding in 2018.  I will also discuss our potential rulemaking geared to help increase transparency for broker order routing and for alternative trading systems, or dark pools.  Before I start, I must provide our standard disclaimer that the views I express today are my own and do not necessarily reflect the views of the Commission, Chairman Clayton, other Commissioners or my colleagues on the SEC staff.^[1]   

Pilots

To start, I would like to pick up from the comments that Chairman Clayton made regarding the 
pilot program for transaction fees that the Commission proposed last month.^[2]   However, I will first make a few comments about pilots generally, before moving on to discussing specific pilots. 

I recently began a review of the collection of pilots that the Commission has authorized and that are being run out of the Division of Trading and Markets.  There are currently 17.  We have to be mindful that there are costs associated with these pilots.  Pilots should be structured efficiently to minimize the direct cost of compliance and the indirect cost that may come from altering the behavior of market participants.

Some of the pilots are specific to an individual exchange.  Many, including Limit Up Limit Down, the Tick Pilot or the Options Penny Pilot, are important forays into new policy areas for the market as a whole where there is generally a lack of sufficient data to enable an empirically based determination of policy.  When important policy issues are on the table, many individuals might think they know the best possible policy outcome based upon their expertise or expertise, however, there often simply is not ample data to persuasively support one policy approach over another.   

That is why it is my view that pilots, when properly constructed, should continue to be a way to make progress on important policy areas without necessarily locking us into a solution that might turn out to be suboptimal.  

Tick Pilot 

As an example of this, I will turn to a pilot currently in operation – the pilot program to study the impact of wider pricing increments in stocks of smaller companies, known as the “Tick Pilot.”  The Tick Pilot provides an interesting case study.  There were those who believed that a wider tick increment would improve the liquidity for smaller companies, potentially leading to more profitability for trading firms, more research analysts at broker-dealers and even more jobs.  There were others who were more skeptical about some of the broader aspirations of such a pilot, but were still interested in whether there may be benefits to the trading of certain smaller securities with a wider trading increment.  In this particular case, the US Congress also got involved, and legislation passed the House of Representatives calling for a pilot.  Without going over all of the process details, the SROs eventually designed a pilot as part of a National Market System Plan, which the Commission approved, and we are now nearly a year and a half into the Tick Pilot.   

The Tick Pilot is set, under its own terms, to expire on October 2 of this year, at which point the Commission will need to decide to either allow it to expire, or whether it should be extended.  On March 29, 2018, the Division of Trading and Markets, acting for the Commission pursuant to delegated authority, issued an exemption that provides the SROs additional time to submit a joint assessment of the impact of the Tick Pilot.  That joint assessment is now due on July 3, 2018.  While I will carefully review the assessment when it is submitted, my belief at this time based on the data currently available and after consultation with DERA is that the best course will be to end the Tick Pilot as scheduled.

This is an example of a pilot that has the ability to provide us with much useful data, but also one that may demonstrate that a policy solution worth exploring is not necessarily a policy solution that makes sense for the long haul.  If the Tick Pilot ends in October, we will still conduct a fulsome analysis of the data generated, along with participants, academics and other interested persons, so that we may fully benefit from the data.  I think there is still much that we can learn from the Tick Pilot, including the relative changes in trading costs associated with a wider spread and the costs and benefits associated with a trade-at provision.  In this respect, the data generated by the Tick Pilot will continue to inform market structure questions on an ongoing basis.  Indeed, I am hopeful that some of these finding could even be useful in our assessment of a potential Transaction Fee Pilot. 

On the Transaction Fee Pilot 

Switching to that topic, I would like to add a few comments to complement the Chairman’s, echoing a few of the points I made to the Commission on March 14 when that Pilot was proposed. 
 
The creation of a Transaction Fee Pilot was one of the most prominent recommendations from the Equity Market Structure Advisory Committee (“EMSAC”), and while there may not be consensus about some of the elements of a pilot, it is one of the few areas where there is significant consensus among market participants that something needs to be done.

A Transaction Fee Pilot will allow the Commission, market participants, and the public to observe how order routing behavior, execution quality, and market quality may change across different test groups, which should further inform consideration of a number of regulatory questions.  

The proposed Transaction Fee Pilot will test not only the effects of lower exchange access fee caps, but also of a prohibition on rebates.  The inclusion of a test group prohibiting rebates is likely the most significant departure from the EMSAC recommendation for a pilot.  Given the comments that we received in connection with the EMSAC recommendation and the related issues at stake, I believe that we would be missing an important and unique opportunity to fully evaluate exchange pricing models without proposing this feature within the Transaction Fee Pilot’s scope.

With this Pilot, we would expect to be able to collect data that will help us better understand how rebates – and rebates at varying levels – impact order routing, liquidity provision, depth on exchange order books, and quoted spreads across varying segments of securities.  With lower – or no – rebates, how will market participants be affected?  For example, would institutional traders be more likely to obtain queue priority and more frequently capture the quoted spread, i.e., buy on the bid, sell on the offer?  Would there be more or fewer retail limit orders displayed on exchanges?  What is the net effect on market quality?  Please keep in mind that the research and commentary provided by all of you is an important part of our ability to evaluate these things.

I would like you to think about where this Pilot could take us.  For example, there are those who believe that the best policy result would be for a tiered access fee cap structure based upon the specific liquidity characteristics of different securities.  In this scenario, there could be a potential outcome whereby the SEC could not only be providing fee caps at a single level but could set fee caps at multiple levels.  This scenario would also require a hard look at the appropriate role of regulation in transaction fee pricing.  

In contrast, for the test bucket without exchange rebates, we would like to better understand whether competitive market forces could successfully cap access fees without a government-imposed cap.  Will the Transaction Fee Pilot data provide insight about what transaction fee pricing would be in the absence of fee-based incentives and, if so, could it provide data about circumstances where a fee cap might no longer necessary?  In the release, we are also asking for comment on whether the public believes that fee caps will always be necessary as long there is an order protection rule within Regulation NMS. 

Transparency Initiatives
 
Before moving on to a more specific discussion of our planned staff roundtables, I would like to make two brief comments on our significant transparency initiatives for broker alternative trading systems (“ATSs”) and for broker order routing disclosure.  First, I found the comments on these proposed rules to be very informative and helpful.  We have given significant consideration to these comments, and I believe that final rules that we hope to move forward have benefitted significantly from your thoughtful input.  I cannot emphasize enough the importance of the public’s role in commenting on these and other initiatives.  

Second, transparency, while generally a highly beneficial principle, can at times be a double-edged sword.  There may be times when too much information about systems design could result in giving away intellectual property.  There may also be times when public disclosure for the sake of disclosure does not always enable fair and accurate apples-to-apples comparisons of order routing behavior and could, instead, lead to a false impression of behavior.  I hope that, as the Commission moves forward with these important rules, we will have arrived at the right balance in these important areas.  

I would now like to discuss the three new staff roundtables that we are planning.  I share Chairman Clayton’s expectation that the roundtable discussions hosted by the Division of Trading and Markets will be targeted on specific issues and lead to actionable recommendations.  To that end, I want to add some detail about the roundtable topics.

Staff Roundtable on Market Structure for Thinly-Traded Securities

The first staff roundtable on April 23 will address the market for thinly-traded securities – both equities and ETPs.  For NMS stocks of all types, one-half have an average daily trading volume of less than 100,000 shares.  Of the approximately 4,656 corporate stocks currently listed on U.S. exchanges, 1301 have daily trading volume of less than 100,000 shares.  These thinly-traded common stocks represent 14.9% of all NMS stocks, but only 0.7% of all NMS trading volume.  Turning to ETPs, 1,617 have average daily trading volume of less than 100,000 shares.  These thinly traded ETPs represent 18.4% of all NMS stocks, but only 0.4% of all NMS trading volume.^[3]  

Today, we have a single equity market structure for all NMS stocks, large and small, liquid and illiquid.  As Chairman Clayton emphasized, we must ask ourselves whether this market structure is appropriate for thinly-traded securities.

The Capital Markets Report from the Department of Treasury highlighted the challenges that face buyers and sellers when a relatively small volume of trading is spread out among many different venues.^[4]   It recommended that the SEC consider allowing issuers of less liquid stocks to suspend unlisted trading privileges, or UTP, for a stock so that it would trade on a smaller number of venues until liquidity reached a minimum threshold.  To maintain a basic level of competition, the Capital Markets Report recommended that broker internalization should remain as a trading option for all stocks.

This recommendation, as well as other ideas and issues, will be evaluated in detail at the staff roundtable.  We will discuss whether limiting UTP would provide a better opportunity for exchanges to innovate to serve thinly-traded securities.  If so, are there ways to address concerns about monopoly exchange trading privileges, as well as the more complex coding issues that might come with connecting to exchanges with different market structures?  Considering the potential benefits of appropriately structured pilot programs I noted above, is there potentially a pilot program that would help the Commission evaluate market structure changes for thinly-traded securities?

Staff Roundtable on Access to Markets and to Market Data

The second staff roundtable that we anticipate would focus on issues related to access to markets and to market data.  Market participants have access to a wide range of products and services that offer differential access to markets and to market data.  This differential access raises important questions, some of which are the subject of proceedings currently pending before the Commission.  Any roundtable would follow and build upon the Commission’s resolution of those questions.   Among the issues that we currently believe such a staff roundtable could explore are: 

• Differing views concerning the sale of market data products and services and its impact on the goal of maintaining fair, orderly, and efficient markets.
• The impact and desirability of many market participants getting data at slower speeds because they are unwilling or unable to pay the rates for the fastest access. 
• The extent to which the markets and investors would benefit from greater transparency into the economics of market data distribution by exchanges and the SIPs, or Securities Information Processors.
• The pros and cons of the existing single consolidator model for SIP data versus a competing consolidator model.
• Differing views surrounding the governance model behind the operation of the SIPs, and whether that model is subject to conflicts of interest that could impede its ability to arrive at the best possible outcomes for investors. 

Staff Roundtable on Regulatory Approaches to Combat Retail Fraud

The third staff roundtable that we anticipate will address regulatory approaches to combat retail fraud.  As Chairman Clayton consistently has emphasized, we must continue to focus on the long-term interests of Main Street investors in everything we do at the SEC, including our efforts to strengthen market structure.  A sound market structure, for example, benefits retail investors in a variety of ways.  It helps assure that their transactions are handled by intermediaries that are trustworthy and will protect investor funds and securities.  A fair and efficient market structure also helps assure that investor transactions are executed at prices that reflect the true forces of supply and demand, rather than being distorted by the effects of fraud and manipulation.  I anticipate that, among other things, the third staff roundtable will address these fundamental market structure objectives in the contexts of digital assets and penny stocks. 

Conclusion

To finish up, I will echo Chairman Clayton’s call for your engagement on equity market structure.  As evidenced by many of the complex market structure issues I have noted today, the full participation of a wide range of investors and other market participants is critical to support an informed and holistic assessment of U.S. equity market structure.  The Transaction Fee Pilot has been proposed and awaits your feedback.  And the primary objective of each of the coming staff roundtables will be to provide the public with a full and early opportunity to weigh in on concrete initiatives that may move equity market structure forward.  I am confident that, with your help, we can achieve this objective.

Thank you