0001
 1           U.S. SECURITIES AND EXCHANGE COMMISSION 
 2   
 3   
 4   
 5   
 6                     EQUITY MARKET STRUCTURE
 7                   ADVISORY COMMITTEE MEETING
 8   
 9   
10   
11   
12   
13   
14                  Tuesday, April 26, 2016 
15                          9:30 a.m. 
16   
17   
18   
19   
20   
21   
22   
23           U.S. Securities and Exchange Commission 
24            100 F Street, N.E., Washington, D.C. 
25              Station Place 1 Multipurpose Room 
0002
 1   PARTICIPANTS:  
 2    
 3   Chair Mary Jo White  
 4   Commissioner Michael Piwowar  
 5   Commissioner Kara Stein  
 6   Steve Luparello  
 7    
 8   Matthew Andresen  
 9   Reginald Browne  
10   Kevin Cronin  
11   David Cushing 
12   Jeffrey Davis 
13   Mark Flannery  
14   Thomas Farley  
15   Gary Goldsholle  
16   Charles Jones 
17   Brad Katsuyama 
18   Ted Kaufman 
19   Richard Ketchum  
20   Manisha Kimmel  
21   Mehmet Kinak  
22   Faris Matalka 
23   Joseph Mecane  
24   Jamil Nazarali  
25   Eric Noll  
0003
 1   PARTICIPANTS (CONT.):  
 2    
 3   Adam Nunes 
 4   Maureen O'Hara  
 5   Joe Ratterman  
 6   Brett Redfearn  
 7   David Shillman 
 8   Nancy Smith 
 9   Chester Spatt 
10   Gary Stone 
11    
12    
13    
14    
15    
16    
17    
18    
19    
20    
21    
22    
23    
24    
25    
0004
 1                        C O N T E N T S 
 2                                                        PAGE 
 3   Call to Order and Opening Remarks                      5  
 4    
 5   Presentation on Regulation NMS Subcommittee  
 6             Recommendations                             22 
 7    
 8   Panel 1:  Consideration and Discussion of  
 9             Framework for Potential Access Fee 
10             Pilot                                       31 
11    
12   Presentation on Trading Venues Regulation  
13             Subcommittee Recommendations                95 
14    
15   Panel 2:  Consideration and Discussion of  
16             Recommendations Relating to  
17             Trading Venues Regulation.                 112 
18    
19   Presentation of Market Quality Subcommittee 
20             Status Report                              146 
21    
22   Presentation of Customer Issues Subcommittee 
23             Status Report                              162 
24    
25   Adjournment                                          170 
0005
 1                     P R O C E E D I N G S 
 2             MR. LUPARELLO:  Good morning.  I would like to 
 3   call the meeting to order.  I believe we have a quorum 
 4   this morning.   
 5             Thank you all today for joining us for the 
 6   fourth meeting of the SEC's Equity Market Structure 
 7   Advisory Committee.  On behalf of the committee and as 
 8   the designated federal officer of the committee, I'd like 
 9   to welcome Chair White this morning and ask her to make 
10   some opening comments. 
11             CHAIR WHITE:  Thank you, Steve, very much, and 
12   good morning and welcome to everybody, members and 
13   panelists, for participating in today's meeting. 
14             Let me actually begin my remarks by just noting 
15   how impressed I am with the engagement and progress of 
16   this committee in tackling the issues that are central to 
17   optimizing equity market structure. 
18             Your agenda for today and the documents made 
19   publicly available by the four subcommittees clearly 
20   reflect thoughtful analysis and recommendations. 
21             The full committee will be considering specific 
22   recommendations from two subcommittees today and 
23   receiving status reports from the other two 
24   subcommittees. 
25             The Regulation NMS Subcommittee and the Trading 
0006
 1   Venues Regulations Subcommittee are both presenting 
 2   recommendations for consideration by the full committee 
 3   today.  They are:  a proposed framework for an access fee 
 4   pilot that would measure the impact of reduced access 
 5   fees on current equity market structure and several 
 6   recommendations for changes to the current trading venue 
 7   rules on SRO immunity; NMS planned governance; changes in 
 8   the SRO proposed rule process when technology changes are 
 9   also required, and the centralization of common 
10   regulatory functions. 
11             All of these recommendations follow independent 
12   analysis by SEC staff available on the committee's web 
13   page; presentations and advice from an array of 
14   panelists; and discussions at prior public meetings. 
15             I'm very pleased to see this progress.  
16   Discussion and consideration of the subcommittees' 
17   recommendations are the next step, which I know will be 
18   well informed by the thorough analysis of the 
19   subcommittee members themselves, the full committee's 
20   views, and by the views of the knowledgeable and diverse 
21   panelists invited to present at the meeting today. 
22             As you know, I'm fully committee to pursuing 
23   market structure reforms that enhance our markets, that 
24   are driven by data and careful analysis.  I think this 
25   committee's current deliberations regarding an access fee 
0007
 1   pilot for the production of data that would inform more 
 2   permanent market structure changes is consistent with 
 3   that principal and approach.  Of course, any 
 4   implementation of any access fee pilot would require both 
 5   commission approval and the significant participation and 
 6   cooperation of the SROs. 
 7             For my part, I believe a properly designed 
 8   access fee pilot would be an appropriate step in 
 9   futhering our collecting assessment of this significant 
10   aspect of our current market structure.  However, as is 
11   evident from the recommendation to be discussed today, 
12   there are a number of significant issues in any access 
13   fee pilot that require close scrutiny and thorough 
14   planning. 
15             I'm particularly interested in hearing more 
16   about the views on certain of the elements of the 
17   recommended pilot.  I'll just very quickly mention four.
18             First, a de minimis bucket of securities 
19   subject to a two mil fee cap is addressed.  What is the 
20   thinking about the possibility of having a group of 
21   securities for which access fees and rebates would be 
22   prohibited all together?  What are the views about a de 
23   minimis approach versus a no-fee and rebate approach?  
24   Would the de minimis approach compromise the goal of 
25   generating data to determine the impact of access fees on 
0008
 1   our market structure? 
 2             Second, I'm interested in hearing discussion 
 3   about whether the pilot should apply to non-displayed 
 4   liquidity in off-exchange trading venues such as 
 5   alternative trading systems.  As the recommendation 
 6   recognizes, Commission Rule 610 only imposes access fee 
 7   caps on exchange-displayed liquidity as market 
 8   participants are prohibited from trading through such 
 9   quotes when they are protected.  What would be the 
10   relevant considerations for expanding a pilot program to 
11   quotes and venues that are not subject to the access fee 
12   cap today?  
13             Third, the recommendation does not include a 
14   trade-at provision, noting that one purpose of the access 
15   fee pilot is to assess whether a trade provision is 
16   needed and further noting that trade-at, while 
17   intertwined with access fees, concerns a different market 
18   structure issues. 
19             I look forward to hearing the perspectives on 
20   this issue and more about the rationale for the 
21   recommended choice. 
22             Finally, to turn to the data-driven nature of 
23   our review, what is the optimal timing for an access fee 
24   pilot? 
25             The primary purpose of an access fee pilot is 
0009
 1   to generate data about impact on liquidity and market 
 2   behavior.  In light of the fact that the tick size pilot 
 3   is scheduled to begin on October 3rd, is there any 
 4   concern that the utility of any data produced from an 
 5   access fee pilot would be degraded if the two pilot 
 6   programs were to run in tandem, or can the access fee 
 7   pilot be designed to avoid that? 
 8             I'm also very interested in today's discussion 
 9   regarding the trading venue regulation recommendations.  
10   But in the interest of time, I won't highlight any 
11   specific recommendations or questions, but I do look 
12   forward to a robust discussion concerning the concrete 
13   measures that are proposed, including how they might 
14   mitigate inherent conflicts of interest. 
15             Later this afternoon, the Customer Issues 
16   Subcommittee and Market Quality Subcommittee will be 
17   providing updates on the significant issues they are 
18   considering.  It's clear from the written reports that 
19   both subcommittees have begun to coalesce around a number 
20   of potential recommendations.  
21             Among other topics, the Market Quality 
22   Subcommittee has been actively reviewing how the U.S. 
23   equity markets operated during the volatility of August 
24   24th and what may flow from what was observed on that 
25   day. 
0010
 1             The Customer Issues Subcommittee is considering 
 2   a range of potential enhancements to our current rules 
 3   requiring disclosure of execution quality and routing 
 4   information with a particular focus on retail investors. 
 5             This work is both critical and timely, as the 
 6   industry and regulators also continue to consider these 
 7   issues and possible changes to existing practices.   
 8             As described in an earlier meeting, for 
 9   example, the Commission staff has conducted an extensive 
10   analysis of the events of August 24, exploring a range of 
11   issues, including the opening and reopening process at 
12   the primary listing exchanges; the effects of market 
13   volatility on trading in certain ETPs and corporate 
14   stocks; the operation of the limit up-limit down pilot 
15   plan; and the specific reasons behind the number of 
16   trading pauses in ETFs. 
17             I'm pleased to see that the Market Quality 
18   Subcommittee is keenly focused on these and related 
19   issues.   
20             As you know, the Commission, late last week, 
21   extended the limit up-limit down pilot period for one 
22   year and specifically recognized the ongoing work being 
23   conducted by the limit up-limit down participants to 
24   further refine the plan's operation.   
25             That work includes considering the 
0011
 1   harmonization of current clearly erroneous execution 
 2   rules with the plan; a review of ETPs to determine 
 3   whether adjustments should be made to the limit up-limit 
 4   down to account for their particular trading 
 5   characteristics; potential enhancements to the 
 6   categorization of securities into different tiers; and a 
 7   review of other issues that may have arisen during the 
 8   events of August 24th, including the impact of double-
 9   wide price bands during the opening period and the 
10   availability of coordinated reopening procedures. 
11             I'm confident that this committee, with the 
12   able engagement of the Market Quality Subcommittee and 
13   the advice of expert panelists, with provide invaluable 
14   input to the Commission and the planned participants on 
15   the limit up-limit down rule and related trading rules. 
16             You'll also hear this afternoon from the 
17   Customer Issues Subcommittee on issues related to 
18   execution quality and order routing disclosures.  As you 
19   know, the Commission staff has been directed to focus on 
20   improvements to our Rule 606 routing disclosures with a 
21   particular focus on institutional order routing 
22   disclosures and discrete enhancements to the existing 
23   retail order routing information provided by broker-
24   dealers today. 
25             I expect the staff's recommendation will be 
0012
 1   considered by the Commission in the very new future, and 
 2   I look forward the Customer Issue Subcommittee's report 
 3   on their consideration of issues in this area. 
 4             So, let me just conclude by saying that today's 
 5   robust agenda really does, I think, demonstrate the 
 6   importance of gathering a diverse set of market 
 7   participants to address complex, potentially very 
 8   significant market structure issues. 
 9             As I've said before, while we do not require 
10   perfect solutions, our regulatory changes must be 
11   informed by clear-eyed, unbiased, and fact-based 
12   assessments of the likely impacts, positive and negative, 
13   on market quality for investors and issuers.  Continued 
14   engagement by this committee and all market participants 
15   on the range of potential issues is truly critical. 
16             So, thank you again for your time and effort 
17   that you're committing to this committee.  Thank you. 
18             MR. LUPARELLO:  Thank you, Chair White.   
19             I will now ask the Commissioners to make their 
20   opening statements, staring with Commissioner Stein. 
21             COMMISSIONER STEIN:  Good morning and welcome 
22   to this meeting of the Equity Market Structure Advisory 
23   Committee. 
24             As the Chair mentioned, today's agenda covers 
25   two recommendations that have come from the hard work of 
0013
 1   two of the Committee's subcommittees.  But first, I want 
 2   to take this opportunity to note a significant item on 
 3   the Commission's calendar for tomorrow.  Tomorrow the 
 4   Commission will consider issuing a public notice on a 
 5   proposed plan submitted by the SROs to create, implement, 
 6   and maintain a consolidated audit trail.   
 7             The development and deployment of a 
 8   consolidated audit trail, which we call the CAT, may be 
 9   one of the most important market structure items on -- 
10   that the Commission undertakes this year.  
11             I want to thank the chair for prioritizing 
12   this, as it is critical for the Commission in fulfilling 
13   its mission to ensure fair and orderly markets. 
14             Turning to the work of the Equity Market 
15   Structure Advisory Committee, I am pleased to see that 
16   we're being presented with two recommendations today.  
17   The first, as the Chair mentioned, is from the NMS 
18   Subcommittee and concerns a framework for a potential 
19   access fee pilot. 
20             As everyone knows, this idea has been floating 
21   around for a number of years, and I want to commend the 
22   subcommittee for tackling this and suggesting a way 
23   forward.  I very much look forward to hearing from 
24   today's panelists and the ensuing discussion. 
25             In particular, I'm interested in learning about 
0014
 1   how we can best obtain a robust data set.  Could this 
 2   pilot incorporate a trade-at provision?  If so, how?  
 3   What are the pros and the cons of doing so?   
 4             Further, I'm also interested in learning about 
 5   potential timelines.  How long would it take to develop 
 6   and to implement?  How long should a potential pilot run? 
 7             The second item on the agenda concerns 
 8   recommendations regarding trading venue regulation.  With 
 9   ever newer and faster technologies, our trading market is 
10   constantly evolving.  The SROs and the Commission need to 
11   evolve, too.  This is critical to the markets remaining 
12   fair and orderly and vital to maintaining a high level of 
13   investor confidence. 
14             As our markets change, our regulatory approach 
15   must also change, which is, in large part, why we have 
16   created this subcommittee and committee.  We need to take 
17   into account new tensions, risks, uncertainties, and 
18   conflicts. 
19             In 2008 our nation learned an important lesson 
20   when our regulatory approach failed to keep pace and 
21   remain effective.  For some time, the Commission's agenda 
22   has been focused on mandated rules and regulations to 
23   address problems revealed during the financial crisis.  
24   However, we must find the time and attention to be 
25   forward-thinking. 
0015
 1             I applaud the work of this Committee that calls 
 2   our attention to how we can optimize our rules to address 
 3   today's concerns, and I look forward to the dialogue with 
 4   the panel. 
 5             Finally, I look forward to the status reports 
 6   of the Market Quality and Consumer Issues subcommittees. 
 7    As I've said in the past, I very much appreciate the 
 8   time and attention that each of you have dedicated to the 
 9   important work of this committee.  This work requires you 
10   to put aside your own business interest and to think 
11   through solutions for our markets as a whole.  And I'm 
12   counting on you; the American people are counting on you 
13   to recommend improvements to our equity market structure 
14   for the benefit of all investors and our nation.   
15             I look forward to today's analysis and 
16   discussion about what is best for our capital markets.  
17   Thank you. 
18             MR. LUPARELLO:  Thank you, Commissioner Stein. 
19             Commissioner Piwowar. 
20             COMMISSIONER PIWOWAR:  Thank you, Steve, and 
21   I'm pleased to join all of you this morning.   
22             I want to take this opportunity to make a few 
23   personal observations about -- for the subcommittee 
24   recommendations that will be discussed today. 
25             The Regulation NMS Subcommittee's 
0016
 1   recommendation for an access fee pilot is provocative, to 
 2   say the least, judging from the reactions that we have 
 3   heard in just the last week.  I'm delighted to see that 
 4   this is the case, because it means that we can expect 
 5   full engagement of market participants to help us vet the 
 6   framework.  I expect today's discussion will be animated, 
 7   to say the least. 
 8             We have been told by many parties, including 
 9   members of Congress, such as Senators Mark Warner and 
10   Mark Crapo, to act prudently with respect to an access 
11   fee pilot.  Indeed, because the key to the success of any 
12   pilot is the specifics of the design, I will always been 
13   keenly focused on whether the design of a particular 
14   pilot is sound. 
15             To that end, let me share my initial reaction 
16   on a single piece of the proposed access fee pilot 
17   framework.  My preliminary view is that it lacks a key 
18   category for observation, namely a zero access fee 
19   category on the basis that the proposed bucket with the 
20   smallest access fee cap will itself result in a de 
21   minimis economic incentive.  And while that may be true 
22   under the Commission's current tick size regime, it may 
23   not be the case if the Commission decides in the future 
24   to lower the minimum tick size.   
25             I hope that the possibility of a zero-fee 
0017
 1   bucket is something we can further explore.  I was glad 
 2   to hear that Chair White agrees with me on that, and I 
 3   encourage public comment on that permutation. 
 4             Immunity for SROs is something that has been of 
 5   significant interest in the marketplace, especially in 
 6   light of several high-profile events in recent years.  
 7   The concept is certainly worth exploring. 
 8             The Trading Venue Subcommittee's recommendation 
 9   will give interested parties a single place to provide 
10   comment rather than having views interspersed in and 
11   among various rule filings, which will facilitate 
12   consideration of this issue.  The NMS planned governance 
13   topics the Trading Venue Subcommittee has identified with 
14   respect to advisory committees are particularly timely.  
15             As Commissioner Stein noted, tomorrow the 
16   Commission will have an open meeting on the consolidated 
17   audit trail.  I hope that you will recognize some of the 
18   subcommittee's insightful work incorporated into 
19   questions in the CAT notice and comment.  The information 
20   we gather in connection with the responses to the CAT NMS 
21   plan can inform how we proceed on plan governance 
22   generally and vice versa. 
23             Speaking of CAT, it has the promise of creating 
24   new and powerful means of surveilling activity that is 
25   conducted on multiple markets.  The Trading Venues 
0018
 1   Subcommittee is suggesting that responsibility for such 
 2   cross-market surveillance and similar common functions be 
 3   centralized in a single regulator.  But I think it's 
 4   worth considering whether there would be benefits to 
 5   having SROs compete with one another for that purpose.  
 6   Public comment will be particularly useful as I think 
 7   through this issue. 
 8             Before I close, let me acknowledge the other 
 9   two subcommittees, the Market Quality Subcommittee and 
10   the Customer Issues Subcommittee.  I understand that 
11   you've been hard at work analyzing issues and have 
12   developed a few preliminary recommendations that will be 
13   presented for consideration at future meetings.  I very 
14   much look forward to hearing your ideas and, time 
15   permitting today, I hope that you will provide us 
16   previews of those coming attractions.  I'd hate to lose 
17   several months simply because of scheduling issues. 
18             I would also be remiss if I did not thank the 
19   panelists and the staff for your contributions to today's 
20   event.  As usual, you play an invaluable role in these 
21   proceedings. 
22             At a prior meeting I challenged the committee 
23   to take ownership of the issues and generate forward 
24   momentum for an equity market structure review, and 
25   you've done just that.  Your work has matured into 
0019
 1   detailed proposals that will spur much-needed public 
 2   debate.  Please, please, please, keep bringing us 
 3   thoughtful recommendations and, in turn, challenge us to 
 4   respond.   
 5             Thank you. 
 6             MR. LUPARELLO:  Thank you, Commissioner 
 7   Piwowar. 
 8             I'd like at this moment to introduce my 
 9   colleagues in the Division of Trading and Markets.  To my 
10   immediate right is Gary Goldsholle, the deputy director 
11   of the division, and to his right is David Shillman, who 
12   is the associate director for the Office of Market 
13   Supervision.  You'll see the three of us standing up in a 
14   room very close to this tomorrow morning on CAT. 
15             Also joining us today at the other end of the 
16   table is Mark Flannery.  Mark is the chief economist and 
17   the director of the Division of Economic and Risk 
18   Analysis, also known as DERA. 
19             Standard disclaimers apply to today's meeting. 
20    Our views expressed in this forum are ours and cannot be 
21   attributed to the Commission or the Commissioners or the 
22   body of the Commission as a whole.  
23             Since our meeting last February, each of the 
24   subcommittees has held several meetings and proactively 
25   solicited participation from various types of market 
0020
 1   participants, including those not directly on the 
 2   advisory committee. 
 3             Although the subcommittees have been self-
 4   directed in terms of defining their agendas and 
 5   conducting their meetings, they have taken care to be 
 6   inclusive by inviting participation from a wide range of 
 7   constituencies and consulting as necessary with other 
 8   advisory committees, including the Commission's Investor 
 9   Advisory Committee and their members.  These 
10   consultations are reflected in the summary minutes of the 
11   subcommittee meetings which are public and posted on the 
12   committee website and factored into the subcommittee 
13   recommendations or status reports that will be discussed 
14   at today's meeting. 
15             Of course, any advice or recommendations of a 
16   subcommittee must and will be presented to and discussed 
17   and considered by the full committee at one or more 
18   public meetings. 
19             Today both the Reg NMS Subcommittee and the 
20   Trading Venues Subcommittee will be presenting their 
21   recommendations to the full committee for consideration. 
22    To set the stage, each chair of the respective 
23   subcommittee will outline and describe their 
24   recommendations.  We'll then hear from the panelists on 
25   the proposed recommendations, and then, we'll turn it 
0021
 1   over to discussion among the committee members, the 
 2   chair, and the commissioners. 
 3             While the Committee may reach sufficient 
 4   agreement and decide to vote on a recommendation, today's 
 5   recommendations involve a number of complex issues, as 
 6   we've already heard, that warrant thorough consideration, 
 7   including significant issues on which subcommittee 
 8   members were not able to reach a single common view.  The 
 9   Committee may decide that these recommendations need 
10   further development after today's discussion among 
11   panelists and other committee members. 
12             If the Committee determined that additional 
13   consideration is necessary, then I would expect that the 
14   subcommittees would meet in the interim to further 
15   deliberate and refine their recommendations.  And when 
16   prepared to present it to the full committee, the 
17   recommendations would be circulated to the committee and 
18   publicly posted. 
19             Depending on the members' schedules and 
20   preferences, a full committee meeting could occur as 
21   early as next month, either in person or telephonically, 
22   to vote on a discrete recommendation or on the current -- 
23   or we could meet on the current quarterly schedule. 
24             And this goes without saying, but throughout 
25   the committee's term, interested parties may also submit 
0022
 1   comments on the work of the committee and its 
 2   subcommittees via the Equity Market Structure Advisory 
 3   Committee website. 
 4             Now that I've gotten that somewhat thorough 
 5   administrative procedure out of the way, let's move on to 
 6   the first agenda item for the day, which is the framework 
 7   for a potential access fee pilot.  I'll turn it over to 
 8   Kevin whose bruises seem to have healed nicely, and I'll 
 9   let him kick it off. 
10             MR. CRONIN:  Thanks, Steve.  And thanks 
11   everybody for the opportunity to present what we think is 
12   a reasonable step forward, which is the access fee pilot. 
13    I think, as we had conversations historically, certainly 
14   in this committee level and, I'm sure a number of us well 
15   beyond this committee, over the years the access fees in 
16   particular seemed to be an issue which have tentacles 
17   into all kinds of other issues within market structure.   
18             As we view access fees, certainly as an 
19   institution, it is perhaps the single biggest issue that 
20   has the consequential things of concern to us like 
21   conflicts of interest and the way our orders are routed 
22   predicated on the way that brokers can profit from the 
23   order routing.  And that may or may not coincide with 
24   achieving the best interest for our clients. 
25             So, this is an issue that, certainly as we 
0023
 1   looked at, was one that we believe needed some sort of 
 2   investigation.  I'm happy to report, as we had the 
 3   conversation again both at the larger committee level and 
 4   certainly at the subcommittee level, it does seem like 
 5   there is enough belief that something needs to happen 
 6   around access fees, which is helpful.  
 7             Now, we're all going to define what that 
 8   something is probably differently.  We're all going to 
 9   define what the real issue is very differently.  But I 
10   think at the core of the pilot design and really what 
11   we're trying to achieve here is to have a data-led 
12   scientific discussion, leaving the editorials and the 
13   visceral reactions to the side as best we can, to really 
14   understand are access fees hurtful; are they harmful; are 
15   they helpful to the formation of liquidity and 
16   efficiencies in the marketplace. 
17             So, at its core, that's what we have tried to 
18   design here, an access fee pilot which is designed to 
19   root out the issues and the understanding.  Hopefully, we 
20   can come to a common nomenclature around what success 
21   looks like, what the variables we want to look like -- or 
22   look at rather to determine success are.  And we've 
23   provided a number of these things within our framework. 
24             Chair White, as you properly have pointed out, 
25   there are a number of issues where we could not find 
0024
 1   consensus.  And our belief was, we were going to put a 
 2   framework out, recognizing that we couldn't get 
 3   consensus, but certainly could set the stage for the 
 4   broader discussion to happen in the marketplace, 
 5   certainly at the exchange level, at the Commission level 
 6   etcetera. 
 7             So, what we're trying to do is have a 
 8   thoughtful, hopefully insightful, piece of a framework 
 9   that we've put together to have the conversation on. 
10             Can I just also add -- and then I'm going to 
11   pass it over to Joe Mecane to actually go to the pilot.  
12   But we did want to be inclusive here.  We thought it was 
13   really important for the credibility of the pilot itself 
14   to bring other participants into this conversation, 
15   right.   
16             So, you see a number of participants in front 
17   of you.  Charles, Faris, and Adam all were participants 
18   not only in the initial discussion but throughout the 
19   entire conversation that was had on the fee.  They were a 
20   vital part of the formation of the ideas and certainly 
21   where we've landed in our conclusions.  So, I thank them 
22   for their effort. 
23             I would also be remiss if I didn't mention that 
24   I did invite our friends at NASDAQ in New York to 
25   participate.  Unfortunately, they decided that they 
0025
 1   weren't going to do that, but we did try to be as 
 2   inclusive as possible in this. 
 3             So, with that, I will -- we'll set that as the 
 4   sort of framework for the framework itself and pass to 
 5   Joe to have the discussion on that on the real issues. 
 6             MR. MECANE:  Sure.  So, thanks. 
 7             So, picking up where Kevin left off, starting 
 8   with the premise that there was general agreement about 
 9   the need to revisit the 30 cent access fee cap, we then 
10   started talking about what's the right way to construct 
11   the pilot to determine what types of changes make sense 
12   if we're going to reconsider the NMS access fee limit. 
13             And so, one premise that we started with was 
14   trying to find the balance between a meaningful pilot and 
15   one that stayed as simple and true to its core mission as 
16   possible, so testing as few variables at one time as 
17   possible. 
18             So, where we landed in terms of the range of 
19   buckets -- because once you start talking about if 30 
20   cents needs to be revisited and what's the correct rate 
21   that we should think about, it's really difficult, 
22   especially given that we've seen other experiments 
23   happening in the marketplace in single venues largely 
24   being inconclusive because they have never been conducted 
25   on a market-wide basis. 
0026
 1             It led us to believe that a prudent pilot would 
 2   need to cover a range of possible outcomes and give us a 
 3   better understanding of how access fees impact both 
 4   liquidity provision and liquidity taking in the 
 5   marketplace. 
 6             So, we landed on a four-bucket proposed pilot, 
 7   one, a control bucket; the second, reducing the 30 cent 
 8   cap to 20 cents; the third, reducing the 30 cent cap to 
 9   10 cents; and then the last, reducing the 30 cap to 2 
10   cents. 
11             Specifically to address Chair White and 
12   Commissioner Piwowar's comments, the two-cent bucket was 
13   a little bit of a compromise bucket.  One area that the 
14   committee could not reach consensus on was whether, in 
15   fact, there should be a no-access fee bucket.   
16             And I'm sure that will be the topic of debate 
17   at today's committee meeting, but the two-cent bucket was 
18   picked in an effort to try to pick a bucket that had 
19   something at the very low end of the access fee level, 
20   made the room for rebates reasonably limited, but also 
21   was sensitive to the fact that the average exchange 
22   revenue capture rate is in the three to four cent per 
23   hundred range.  And a two cent per side charge largely 
24   keeps the net economic impact to exchanges similar to 
25   what is today. 
0027
 1             And so, we had that discussion around, you 
 2   know, the pros and cons of completely eliminating 
 3   rebates.  We put two cents in as kind of a compromise 
 4   position, but there was clearly a strong view by some 
 5   members of the committee that we should, in fact, try a 
 6   no-access fee bucket. 
 7             One of the other provisions that we suggest is 
 8   that the coverage universe for the access fee experiment 
 9   starts with stocks with greater than a $3 billion market 
10   capitalization.  That addresses, I think, a comment by 
11   Chair White that we not have a potential access fee pilot 
12   overlap with the tick pilot.  And so, we drew that line 
13   in the sand so that the two could potentially run 
14   simultaneous and not impact each other. 
15             Hitting on a few other highlights and areas 
16   where there wasn't necessarily consensus, another 
17   significant area where we did not reach consensus was 
18   around whether the proposed caps should in some way apply 
19   to inverted venues, venues that pay the takers and charge 
20   the makers.   
21             And the fact that the current wording of Reg 
22   NMS does not apply to inverted venues, I would say, you 
23   know, caused us to split the discussion in terms of 
24   whether we would be introducing a new variable by 
25   applying the cap to inverted venues versus whether it 
0028
 1   would potentially impact the pilot if flow migrated to 
 2   inverted venues because they were not subject to any 
 3   access fee cap. 
 4             And so, I would say that was -- you know, other 
 5   than the no-rebate bucket, probably the second most 
 6   significant area that we didn't reach consensus on. 
 7             Addressing another one of Chair White's points, 
 8   in terms of application to ATSs, we did have that 
 9   discussion.  We do include that as a consideration in our 
10   recommendation, but we do not strongly recommend it, 
11   mainly for the reason that Reg NMS currently -- and the 
12   access fee cap currently do not apply to ATSs.  A lot of 
13   ATS transactions are more commission-like in nature as 
14   opposed to access fee in nature.  And we highlight the 
15   fact that we think introducing this additional variable 
16   to the framework could introduce other biases that don't 
17   currently exist in the marketplace. 
18             And then, two last points that I'll mention, 
19   one which Commissioner Stein and Chair White both 
20   mentioned around the trade-at topic, which was clearly a 
21   very significant part of our discussions.  And we do not 
22   endorse trade-at as part of the access fee pilot, but to 
23   expand on that, I wouldn't say that our lack of inclusion 
24   is necessarily a vote one way or the other on trade-at 
25   itself. 
0029
 1             The main reasons why we don't include trade-at 
 2   are, first of all, we believe that a trade-at provision, 
 3   by its own definition, would force order flow back onto 
 4   exchanges, which we're not drawing an opinion on directly 
 5   but we believe would predetermine an outcome that would 
 6   impact the ability to measure what the intent of the 
 7   access fee pilot is itself. 
 8             The second reason is because we believe trade-
 9   at, aside from just access fees, introduces other 
10   variables into the discussion, specifically counterparty 
11   segmentation and other considerations that a trade-at 
12   rule would impact outside of just the core of the access 
13   fee pilot. 
14             Third reason is we believe that the tick pilot 
15   already includes a provision for trade-at.  And to the 
16   extent that we are measuring what the impact of the 
17   trade-at rule would be in the marketplace, we already 
18   have that experiment underway. 
19             Additionally, going back to the initial point I 
20   made, we did try to keep the proposal as streamlined and 
21   single-variable as possible and specifically wanted to 
22   try to ensure a simple implementation and one that did 
23   not drag out for long periods of time.  And as we could 
24   see from the current tick pilot discussion, trade-at 
25   introduces a significant level of complexity to the 
0030
 1   discussion which we were trying to avoid in the framework 
 2   that we're proposing. 
 3             The last point I'll just make -- or last two 
 4   points, which I said before the last point, is, we do 
 5   mention that, while we don't believe this proposal has a 
 6   significant direct impact on the options market, it is 
 7   worth ensuring that we're not proposing something that 
 8   has spillover effects to other markets.  
 9             And the last last point is that we do also 
10   highlight the fact that, in an effort to try to minimize 
11   unintended consequences, because the current Reg NMS cap 
12   does not apply to non-displayed liquidity, given that the 
13   marketplace currently operates in a framework where 
14   displayed and non-displayed liquidity are treated largely 
15   equivalently, we would advocate that that be potentially 
16   more specifically brought into the pilot so that we would 
17   not end up in a potential situation where you had a low 
18   access fee bucket and a different rate for non-displayed 
19   liquidity. 
20             So, I'll leave it there.  I'll invite, if I 
21   missed anything, any of my fellow subcommittee members to 
22   weigh in with any other points.  But thank you for the 
23   time. 
24             MR. LUPARELLO:  Thank you, Joe.  Thank you, 
25   Kevin. 
0031
 1             Let me quickly introduce our panelists and then 
 2   ask them to offer their views.  Going from my left to my 
 3   right, we start with Jeff Davis, who is the vice 
 4   president and deputy general counsel of NASDAQ.  Next to 
 5   him is Tom Farley, the president of the NYSE Group.  Next 
 6   to Tom is Charles Jones, the professor of Finance and 
 7   Economics at the Columbia Business School, and next to 
 8   him is Faris Matalka, who is the vice president of 
 9   Trading and Order Management at E-Trade.    And finally, 
10   at the end is Adam Nunes, who's the head of Business 
11   Development at Hudson River Trading.   
12             Jeff, can we start with you? 
13             MR. DAVIS:  Thanks, Steve. 
14             Good morning, Chair White, Commissioners, 
15   committee members, and my fellow panelists.   
16             As Steve said, my name is Jeffrey Davis.  I am 
17   vice president and deputy general counsel of NASDAQ, Inc. 
18    I've spent 20 years at the NASD and NASDAQ, supporting 
19   our U.S. regulated markets, including three equities and 
20   options markets here at the SEC; the NASDAQ futures 
21   market at the CFTC; and eSpeed, NASDAQ's trading platform 
22   for U.S. treasuries. 
23             On behalf of NASDAQ, I want to thank the 
24   Commission and the advisory committee members for the 
25   invitation to appear as a guest here today. 
0032
 1             The topic of equity market structure is 
 2   critically important to issuers and investors.  It's also 
 3   important to NASDAQ, resting as it does, at the heart of 
 4   NASDAQ's flagship market, a market that has brought 
 5   pioneering companies like Intel, Apple, Google, and 
 6   Amazon to the public through initial public offerings 
 7   that were conducted on NASDAQ.   
 8             These companies and hundreds others like them 
 9   use the capital they raise on NASDAQ to bring out cutting 
10   edge products that are integral to our daily lives.  We 
11   support them as they grow, and they, in turn, support the 
12   U.S. economy by creating millions of jobs. 
13             Without unduly belaboring the point, NASDAQ, 
14   the New York Stock Exchange, and Schwab wrote a letter to 
15   Chair White, questioning the lack of diversity on this 
16   committee.  This body is charged with assessing and 
17   examining market structure that exists primarily to serve 
18   issuers and investors.  And yet, it lacks a 
19   representative of retail investors; it lacks a 
20   representative of a publicly traded non-financial 
21   operating company; and it lacks a representative of the 
22   two largest markets that list those companies. 
23             NASDAQ appreciates the hard work of the Trading 
24   Venues Regulations and Regulation NMS subcommittees that 
25   brought forth the preliminary recommendations that are 
0033
 1   under consideration here today.  And yet, we think that 
 2   the preliminary recommendations reveal what the Committee 
 3   has lost by lacking true issuer representation.   
 4             None of the preliminary recommendations 
 5   specifically aims at improving market structure for 
 6   emerging stage high-growth companies that fuel the U.S. 
 7   economy.  The preliminary recommendations don't appear to 
 8   improve equity market structure for companies that are 
 9   tapping the public equity markets for the first time or 
10   secondarily, and the recommendations do not appear to aim 
11   at improving market structure for issuers or holders of 
12   less liquid securities or improving the way that those 
13   securities trade. 
14             The 2010 concept released on equity market 
15   structure signaled the need to consider big themes and 
16   issues in equity market structure, including better 
17   serving investors of all kinds, retail and professional, 
18   short-term and long-term.   
19             Tom Whitman of NASDAQ previously testified that 
20   we should think big and change the one-size-fits-all 
21   approach to equity market structure and attempt to 
22   address different types of issuers, different character 
23   of trading. 
24             Yet, the subcommittee recommendations here 
25   today do not reflect these elements of these previous 
0034
 1   recommendations.  They accept the one-size-fits-all 
 2   equity market structure and change different elements of 
 3   a certain part of that market structure that already 
 4   appears to serve as subset of issuers and investors 
 5   extremely well. 
 6             And in the absence of the two largest U.S. 
 7   equity exchanges, what do the subcommittees recommend 
 8   changing to help U.S. investors?  They recommend 
 9   limiting, restricting exchange access fees, increasing 
10   exchange liability, burdening exchange rule filings, and 
11   restricting exchange self-regulation. 
12             NASDAQ is surprised to see these as -- atop the 
13   priorities list for the Equity Market Structure Advisory 
14   Committee.  These are important topics, but NASDAQ 
15   believes that the lack of diversity on the committee is 
16   reflected in these recommendations. 
17             Nonetheless, because the preliminary 
18   recommendations touch upon important topics and the 
19   subcommittees have clearly spent significant time and 
20   energy, we intend to respond fully and thoughtfully in 
21   writing in the comment letter shortly.  However, five 
22   days' notice was insufficient for us to fully consider 
23   these important topics and give them the attention that 
24   they deserve.   
25             Therefore, though we're not fully able to 
0035
 1   respond here today, we hope that our forthcoming comment 
 2   letter will move the dialogue forward.  Thank you. 
 3             MR. LUPARELLO:  Thank you, Jeff. 
 4             Tom? 
 5             MR. FARLEY:  Thanks, Steve, and thanks to the 
 6   members of the committee.  I noted from the minutes that 
 7   were posted on the website, all the meetings that have 
 8   transpired and all the hard work.  I mean, this is an 
 9   all-star team of our equity markets, and I appreciate and 
10   the New York Stock Exchange appreciates everything that 
11   you're all doing. 
12             And just because I don't get the opportunity 
13   too often, I also want to say to Chair White, 
14   Commissioner Piwowar, Commissioner Stein, Steve, Dave, 
15   Gary, the whole team, that you guys are really doing a 
16   terrific job.  Although my colleagues don't necessarily 
17   all agree, I think of, you know, Reg SCI as increasing 
18   robustness.  They agree, but it was a lot of sleepless 
19   nights to get there.   
20             But also, the time stamp initiative, increasing 
21   transparency for investors, the many enforcement actions 
22   that you've brought and/or investigated, again, we 
23   appreciate it, and you're doing a really nice job. 
24             Thank you very much for having me again here.  
25   I'm here today on behalf of the New York Stock Exchange 
0036
 1   and, importantly, the 2600 companies that list on the New 
 2   York Stock Exchange whose interests are not represented 
 3   on this committee. 
 4             The listed companies at the NYSE are 
 5   responsible for nearly every innovation that you can 
 6   think of over the last 225 years.  These innovations have 
 7   improved the lives, even saved the lives of American 
 8   citizens and global citizens.  Henry Ford's mass 
 9   production of the vehicle allowed ambulances to be -- to 
10   replace skittish horses.  Thomas Edison's mass production 
11   of the light bulb provided a more efficient means to 
12   light the way.   
13             The equity markets are first and foremost a 
14   capital formation mechanism that must fuel continued 
15   innovation, job growth, and quality of life improvements 
16   for the next 225 years.   
17             To put it bluntly, the markets are failing 
18   these listed companies.  The listed companies are voting 
19   with their feet.  The number of listed companies in this 
20   country is down by more than half over the past 15 years. 
21             I speak to listed company CEOs, CFOs, investor 
22   relations officers every single day, and they have simple 
23   but powerful expectations for the capital markets.  
24   First, they wish for transparent trading of their stock, 
25   and second, they expect that the SEC and exchanges will 
0037
 1   regulate the trading of their securities to ensure a 
 2   level playing field that roots out and punishes rule 
 3   breakers. 
 4             Incomprehensively, I am here to address the 
 5   recommendations of a subcommittee of a committee that 
 6   does not represent listed companies or listing venue 
 7   viewpoints.   
 8             And if I can just pause and mention Kevin's 
 9   comment that we were invited, indeed we were, and I very 
10   much appreciate it, that we were invited to participate 
11   in the subcommittee. 
12             And just to provide a little bit of context, 
13   there's many -- there are several subcommittees.  Kevin 
14   was most gracious, invited us to participated.  Other 
15   subcommittees did not.   
16             We were invited to participate in another 
17   subcommittee for one meeting in particular, and we've 
18   said all along, including my very first time in front of 
19   this committee, that we think these matters are so 
20   important and the representation on this committee is 
21   fairly skewed, and so those meetings should be held in 
22   the light of day, should be held in a public forum such 
23   as this, should not be held in a dark room or in a cloak 
24   and dagger fashion.  And so, we said we would always 
25   participate in these committees if invited.  But again, I 
0038
 1   thank Kevin for that, but I wanted to comment on that. 
 2             Equally disturbing, the recommendations of the 
 3   subcommittees are aimed towards increasing the profits of 
 4   financial intermediaries while assuring that more trading 
 5   is done in the dark and that the regulatory authority of 
 6   exchanges is reduced.  These recommendations come after a 
 7   year in which many of the ATSs that host our trading have 
 8   been fined for deceptive practices and other unseemly 
 9   behavior. 
10             In addition, as I speak right now at this 
11   moment, we are in the middle of the month with the 
12   highest percentage of dark trading in history.  These 
13   trends should serve as an industry-wide wakeup call that 
14   we need to refocus on the needs of the listed company.   
15             When we received the recommendations of the NMS 
16   Subcommittee and Trading Venues Regulation Subcommittee 
17   just last week, we were very disappointed.  Instead of a 
18   balanced set of perspectives and recommendations that 
19   would bring our industry together, the subcommittees and 
20   the recommendations they produced will pit segments of 
21   the industry against one another and further reduce the 
22   likelihood of anything ever changing. 
23             I want to point out that, after acquiring the 
24   NYSE, Intercontinental Exchange, which is the parent 
25   company of NYSE Group -- and I was involved personally as 
0039
 1   well -- set out to forge a compromise among a cross-
 2   section of market participants.  That wasn't all good for 
 3   exchanges; it wasn't all good for broker-dealers; and 
 4   frankly, it wasn't all good for investors, but it would 
 5   have been better for the broader market.   
 6             Our proposal even included a balanced approach 
 7   to reducing access fees, mere remnants of which are left 
 8   in this latest access fee pilot recommendation of the 
 9   subcommittee. 
10             Most importantly, our balance proposal was 
11   constructed around simultaneously addressing the 
12   unintended but obvious consequences of making economic 
13   changes in isolation. 
14             Some of the members of this committee led a 
15   similar effort in the past that we weren't involved with. 
16    While those efforts were ultimately thwarted by self-
17   interest or other reasons, we hope more progress could be 
18   made by this subcommittee. 
19             Given the insufficient amount of time we have 
20   had to analyze the subcommittee's recommendations, we, 
21   too, will be submitting our views for the public file at 
22   a later date.  However, in response to requests from 
23   Commission staff, I will highlight a few issues for 
24   consideration, specifically with respect to the access 
25   fee pilot. 
0040
 1             Today many -- I'm going to go a little bit 
 2   slowly through this just kind of section here just to 
 3   make sure that our thoughts are clear on kind of the 
 4   logic of this access fee pilot.   
 5             Many exchanges provide a payment or rebate to 
 6   those parties that leave resting bids and offers on the 
 7   exchange.  This anticipated payment is considered by 
 8   market participants when posting orders on displayed 
 9   markets and ultimately therefore leads to better prices 
10   and therefore tighter bid/ask spreads than would have 
11   otherwise been displayed without a rebate. 
12             The access fee pilot proposal would reduce 
13   greatly or even eliminate the payment of rebates.  In 
14   some buckets it would go to zero likely, which would 
15   mean, without question, that the bid/ask spread on 
16   exchanges would widen.   
17             This is particularly important because the 
18   displayed bid/ask spread establishes the price range for 
19   trades in dark venues.  And when displayed spreads are 
20   wider, it would result in more dark trading. 
21             Simply stated, this access fee pilot will not 
22   only ensure that there will be more dark trading but also 
23   that there will be worse prices on exchanges and off-
24   exchange.   
25             I have shared this theory with literally dozens 
0041
 1   of market structure experts throughout the industry, 
 2   including members of the committee, and not one person 
 3   faults this logic.   
 4             Some people argue that dark venue participants 
 5   will have less incentive to trade on dark venues for fee 
 6   avoidance reasons because the fees on exchanges will now 
 7   be lower as part of this pilot.  But then, nearly all 
 8   will concede that this effect would be negligible or non-
 9   existent because dark pools will continue to offer very 
10   low-cost executions and enjoy less stringent regulatory 
11   requirements than exchanges.  
12             The fact of the matter is that mandating lower 
13   access fees without considering how to incent 
14   transparent, regulated trading will result in a much 
15   worse marketplace for listed companies and institutional 
16   and retail investors. 
17             The real conversation we should have as an 
18   industry is to identify not just how to lower the cost of 
19   trading for certain segments of participants but also to 
20   identify how we can improve the outcome for investors and 
21   capital formation for listed companies.  A healthy public 
22   market requires those prices to be displayed to all 
23   market participants and investors.  It is this 
24   transparent and public process that establishes the 
25   prices around which every other aspect of the secondary 
0042
 1   market is built. 
 2             To encourage that process, we believe those who 
 3   take the risk of stating their intentions aloud should be 
 4   rewarded for it, seems basic to us, whether they are 
 5   market makers, institutional investors, retail investors, 
 6   or others, by receiving execution priority over all other 
 7   market participants. 
 8             When I was invited around about a year ago to 
 9   be a part of a panel sitting at this very table, I was 
10   pleased to hear that a majority of my panel, which was 
11   composed of academics and market participants agree with 
12   these common sense views, and they recommended that this 
13   committee consider a plan to encourage transparent price 
14   formation on regulated exchanges, commonly referred to as 
15   "trade-at".  However, these views, like the views of the 
16   listed companies, were discarded by the subcommittees 
17   where these recommendations were conjured up.   
18             If the SEC proceed with this ill-conceived 
19   access fee pilot, public company stocks will have less 
20   displayed liquidity, and spreads will widen at a time 
21   when off-exchange trading is already at record high 
22   levels.  We already know what will happen.  We have 
23   examples.   
24             If you look at the inverted markets, so-called 
25   taker-maker, they trade with wider spreads.  If you look 
0043
 1   at the NASDAQ access fee pilot that they did last year, 
 2   it resulted in wider spreads. 
 3             These entirely predictable results will mean 
 4   higher costs for investors buying or selling securities, 
 5   and these higher costs go, in part, into the pockets of 
 6   market professionals, many of whom were on the drafting 
 7   subcommittee. 
 8             If I could just speak, I have one page -- just 
 9   because I was not going to be here this afternoon, so 
10   it's not too long, starting here on trading venue 
11   regulation.  
12             Would that be okay, Steve?   
13             MR. LUPARELLO:  Sure. 
14             MR. FARLEY:  I'll do so expeditiously. 
15             Because I will not be participating in this 
16   afternoon's meeting, I will briefly comment on the 
17   recommendations of the Trading Venue Regulation 
18   Subcommittee. 
19             It is unsurprising that a group composed 
20   principally of representatives of ATSs and other off-
21   exchange venues or owners of ATSs and off-exchange venues 
22   would recommend a set of proposals to increase their 
23   competitors' costs and liabilities.  But I must admit, 
24   even I was surprised that after a litany of enforcement 
25   actions against ATSs, the subcommittee so far has 
0044
 1   literally made no mention of ATSs or of additional 
 2   disclosure and reporting requirements for these dark 
 3   pools and internalizing broker-dealers. 
 4             In addition, in the spirit of airing conflicts 
 5   real or perceived, it's worth noting that many of the 
 6   subcommittee members belong or -- and fund the trade 
 7   association that's taking legal action over fee changes 
 8   made by the NMS plan to which we a party and to which 
 9   they are now seeking to change the governance according 
10   to these new recommendations. 
11             These subcommittee members' firms profit from 
12   using the data disseminated by these NMS plans.  
13   Accordingly, reducing the fees they pay for that data 
14   would increase their profits. 
15             So, like the access fee pilot proposed by 
16   another subcommittee, the recommendations of this 
17   subcommittee represent, in part, at a minimum, self-
18   interest of the majority of subcommittee members and 
19   certainly not the interest of public companies or 
20   investors. 
21             Concluding, at the New York Stock Exchange, we 
22   believe that this committee was constituted in hopes of 
23   receiving expert technical advice as the Commission 
24   pursues holistic, thoughtful NMS reform that takes into 
25   account a broad swath of views of the industry.   
0045
 1             We have been critical of the composition of 
 2   this committee for excluding listed company, listing 
 3   venue, and retail investor views.  We have also been 
 4   critical of the subcommittee's opaque process.  But we 
 5   remained hopeful that the group would dispassionately 
 6   advice the commission on critical issues affecting end 
 7   investors in listed companies. 
 8             Unfortunately, instead of focusing on how to 
 9   simplify markets and reduce real conflicts in the market, 
10   some on the committee have focused on how their own firms 
11   can increase profits and/or reduce the profits of those 
12   firms not represented without focusing, again, on the 
13   listed company. 
14             The NYSE remains committee to helping the 
15   Commission seek balanced recommendations that improve 
16   market quality for all market participants.  We also 
17   remain committed to the idea that an equity market 
18   structure advisory committee is an effective means of 
19   considering market structure reform. 
20             Therefore, we urge the SEC to not give up but 
21   rather to start anew with a representative committee.  
22   The future of our equity markets is critically important, 
23   and it's not too late to formally include the interest of 
24   issuers in the process. 
25             Thank you. 
0046
 1             MR. LUPARELLO:  Thank you, Tom.  
 2             Charles? 
 3             MR. JONES:  Thank you, Steve.  Thank you, 
 4   Kevin, for the opportunity to participate in this 
 5   subcommittee.  And thank you to Chair White and the 
 6   Commissioners and the staff of the Commission and to the 
 7   other committee members for the opportunity to 
 8   participate this morning. 
 9             I'm a finance professor at Columbia Business 
10   School, and I do research on securities market structure. 
11    Recently, I've written on the design of regulatory 
12   pilots, and so, I'd like to really focus on those design 
13   issues in my introductory remarks today. 
14             It's always important to remember if we're 
15   thinking about a regulatory change like this, what are we 
16   trying to get at; what is the market failure that is 
17   behind it.  And as Kevin mentioned earlier, I think one 
18   of the principal market failures -- potential market 
19   failures anyway -- is the conflict between brokers and 
20   their customers.  Certainly, there are ways that brokers 
21   and customers can resolve that conflict, but it's 
22   certainly possible that that conflict -- that agency 
23   problem could continue, and it's certainly possible that 
24   we would need regulation to fix that problem. 
25             The key for me in designing a pilot to address 
0047
 1   this is to design something that is simple, minimizes the 
 2   cost to implement, and cannot be gamed or avoided so that 
 3   we can actually get actually information about the 
 4   effects of these particular access fees. 
 5             Such a pilot must also have sufficient 
 6   statistical power.  It must last long enough, and it must 
 7   apply to enough stocks so that economists are likely to 
 8   detect the effects that are there using standard 
 9   statistical techniques. 
10             Simplicity is really why we have decided not 
11   to, you know, include trade-at here.  I think the tick 
12   size pilot has mixed tick sizes in trade-at, and you can 
13   see the complexity that has resulted in the tick size 
14   pilot. 
15             One of our goals here is to have something that 
16   is much simpler.  That's also why we've only suggested a 
17   few buckets and a reasonably long pilot, so that market 
18   participants will find it worthwhile to re-optimize 
19   taking the new restrictions into account. 
20             As I think Joe mentioned earlier, we're 
21   recommending that this only apply to stocks over $3 
22   billion in market cap so that it can run separately from 
23   the tick size pilot so there's no overlap in the universe 
24   of stocks that this would apply to. 
25             And while we don't make a final recommendation 
0048
 1   in the document so far, I believe that we need a 
 2   stratified random sample, that this needs to apply to 
 3   active stocks, inactive stocks, high-priced stocks, low-
 4   priced stocks, large cap stocks, mid cap stocks in order 
 5   to see the effects across a heterogeneous set of 
 6   securities. 
 7             One outstanding issue, as has been discussed, 
 8   is the scope of the limits on the fees and/or the 
 9   rebates.  I realize -- you know, I'm not a lawyer, but I 
10   realize the easiest thing to do here is to just change 
11   the access fee cap associated with protected quotes.  And 
12   inverted exchanges and ATSs, as people have mentioned, 
13   would be unaffected by such a change. 
14             But this will be insufficient if you believe 
15   that the main market failure is the agency problem 
16   associated with fees and rebates throughout the system, 
17   not just the fees associated with accessing a protected 
18   quote. 
19             So, if the goal is to understand the agency 
20   problems throughout our equity ecosystem, the limits on 
21   fees and rebates need to apply to as many trading venues 
22   as possible, ideally including ever ATS.  Otherwise, 
23   there are going to be possibilities for circumventing the 
24   access fee limits. 
25             For example, in order to preserve a particular 
0049
 1   rebate stream, brokers might route their order flow to 
 2   other venues that are excluded from the pilot.  So, I 
 3   would advocate in general -- you know, leaving aside all 
 4   the legal considerations, I would advise that this pilot 
 5   should apply as broadly as possible. 
 6             Finally, a quick note on what we should expect 
 7   to see going into this.  As I noted, we should be on the 
 8   lookout for changes in order routing behavior.  That 
 9   should be the easiest thing for us to observe. 
10             In terms of market quality, I don't know what 
11   to expect.  That's really part of the reason that we want 
12   to do such a pilot, all right, is to really see the 
13   effect on market quality. 
14             I expect those changes to be quite modest, so I 
15   don't think we have much chance here of sort of upsetting 
16   the apple cart in a huge way. 
17             But there is one mechanical effect, and I think 
18   Tom mentioned that, and I want to highlight it as well, 
19   which is that gross bid/ask spreads are likely to widen. 
20    If we reduce access fees, you know, that's going to 
21   change how people are going to quote.  They're not going 
22   to receive that liquidity rebate, and they're likely to 
23   widen their bid/ask spreads as a result. 
24             The right way to measure bid/ask spreads is on 
25   a net basis, and so we should basically not be surprised 
0050
 1   to see this nor should we conclude from a widening of 
 2   bid/ask spreads that that's necessarily a worsening in 
 3   market quality.  That's one of the reasons that the 
 4   subcommittee came up with a wide variety of measures to 
 5   see what the impact would be on market quality, looking 
 6   at a much broader set of market quality measures. 
 7             So, I think that's all I'd like to say at this 
 8   point, but I -- you know, I think that really simplicity 
 9   was one of our guiding lights in designing this.  And the 
10   goal is to try to have something that could be put into 
11   place quickly and simply and that would generate a lot of 
12   useful information about the impact of these particular 
13   fees. 
14             MR. LUPARELLO:  Charles, thank you. 
15             Faris. 
16             MR. MATALKA:  Thank you.  Unlike Tom, I only 
17   have one here, so -- single-sided. 
18             Thank you, Chair, Chairman White and the 
19   Commissioners and the staff members for the opportunity 
20   to present in this forum.  I do want to thank Kevin as 
21   well for the outreach to the retail community and the 
22   inclusion on these various discussions, including this 
23   access fee. 
24             I do want to take this opportunity and 
25   highlight and underscore the great benefits that the 
0051
 1   retail investor are currently enjoying in execution and 
 2   in overall trading experience as a direct result of this 
 3   market structure environment. 
 4             Specifically, if you look at the quoted spread 
 5   and the effective spreads -- are at an all-time low.  The  
 6   displayed size is almost double where it was in 2005.   
 7   These are all great data points that benefits the retail  
 8   investors. 
 9             So, as brokers continue to source liquidity and 
10   optimize routing to dealers that offer the highest price 
11   improvement over the NBBO that's already tight and to 
12   dealers that can price-improve more often, these are the 
13   execution quality in terms of our measurements -- and 
14   it's also disclosed on our website.  It's at a all-time 
15   high. 
16             So, in addition to the execution, I think it's 
17   important also to emphasize the overall trading 
18   experience for the retail investor.  Commissions remain 
19   low.  They haven't been raised in years.   
20             Also, as competition grows between brokers, 
21   it's important to emphasize that we serve the retail 
22   investor, and we allow them to trade on their own terms. 
23    And by that I mean by offering market -- the free market 
24   data, streaming market data, level one, level two, mobile 
25   access.  This is a growing portion of channels that we 
0052
 1   receive orders.  These are all technological expenses 
 2   that firms have to bear and -- to increase the 
 3   competition that also benefits the retail customer.  
 4   Also, free education is a big part of what we offer as 
 5   well to the retail investor.   
 6             We acknowledge, obviously, you know, the market 
 7   structure is always a discussion, and then there's always 
 8   room for enhancement.  And we were acutely aware of all 
 9   the discussions that are happening in terms of, you know, 
10   dark volume versus lit volume on the exchanges, SIP 
11   versus direct feed.  We're definitely aware of these 
12   discussions and we encourage data-driven approaches that 
13   the Commission has laid out, and we commend the 
14   Commission for doing that. 
15             We -- so specifically for the proposal that's 
16   on the table, we support a concept of assessing the 
17   access fee on the market with the thought in mind that 
18   there is a clear definition of what success means and 
19   what defines the framework of success -- or failure for 
20   that matter.  
21             We -- so, the variables that are measured or 
22   that propose to be measured in this proposal is -- 
23   there's a slew of them.  And I think some of them answer 
24   to a different question.   
25             And I think, you know, it's important to 
0053
 1   identify what the objective is.  Is the objective here to 
 2   -- is the question here, does liquidity need an 
 3   incentive.  Is that the main objective of this pilot.  
 4   And to that end, if that is the case, we support a small 
 5   -- similar to the NASDAQ experience, a small scale pilot 
 6   that potentially includes 15, 20 stocks in each bucket 
 7   that are liquid stocks that can measure the impact.  And 
 8   we do agree that it has to be applied on all the 
 9   exchanges to get a meaningful impact, a holistic impact. 
10    So, we're -- we are for a proposal that can assess that 
11   impact on a much smaller scale. 
12             If you think about deconstructing a system, a 
13   pricing system that incentivizes liquidity that's been 
14   prevalent in the market -- and it is today, for 15, 18 
15   years -- done on a large scale will have unintended 
16   consequences and potentially could widen the spread and 
17   could potentially be a detriment to the retail investor. 
18             Thank you. 
19             MR. LUPARELLO:  Okay, Faris. 
20             Adam. 
21             MR. NUNES:  Thank you to the Commission staff 
22   and the committee for inviting me to speak. 
23             So, I'm coming at this from the prospective of 
24   a proprietary trading firm.  You know, there were 
25   mentions of conflicts of interest between brokers and 
0054
 1   their customers.  We don't have that.   
 2             When we think about rebates, you know, access 
 3   fees, and the spread, we think in terms of the net of 
 4   those.  And you know, when we think about a pilot, we 
 5   would expect that, if the rebate comes down, that the, 
 6   you know, spread would widen slightly to account for 
 7   that.   
 8             You know, I think, to Faris' point, one of the 
 9   things that will be important to watch is, for retail 
10   investors, did they see increased price improvement kind 
11   of come along with that widening of the spread to get 
12   them back to where they are today. 
13             And it's important to note that the combination 
14   or the net of the spread and the fee effectively balance 
15   the supply and demand for liquidity provision and 
16   liquidity excess.  To me, that is one of the most 
17   important aspects of this pilot, is to see what happens 
18   with that, you know, does it affect displayed size; does 
19   it affect execution costs. 
20             You know, Tom mentioned -- and I think it's an 
21   important point that the rebates that exchanges pay have 
22   the effect of creating artificially narrow spreads.  
23   That's a tool the exchanges have used to compete with 
24   ATSs and internalizers.  That was an important 
25   consideration of the committee.  We opted for something 
0055
 1   that was a bit more surgical and just went after the 
 2   parameter that is currently in the rule.  But that's 
 3   obviously an important concern. 
 4             When I think about this, I think in terms of 
 5   what we see over the counter, is some portion of fee 
 6   avoidance where a lot of dark pools are -- their value 
 7   proposition is avoiding the 30 cent access fee.  And I 
 8   kind of see countervailing effects between a widening 
 9   spread, which might make it more attractive to trade OTC 
10   and fee avoidance.  And I don't know which way that will 
11   come out as far as what happens on the counterverses on a 
12   change, which, frankly, is why I view this as a good 
13   pilot.   
14             I'm a pretty opinionated person, and I tend to 
15   express those opinions.  I think the areas where we don't 
16   know what the outcome is likely to be is a good area for 
17   a pilot so we can see. 
18             And as a firm that supported a data-driven 
19   approach to market structure policy changes, you know, we 
20   view getting the data as a critical step in that process. 
21    So, we view this as a positive step forward in assessing 
22   whether we should make a change to the access fee cap. 
23             I think the other thing to consider -- and this 
24   address the questions on a trade-at provision, is the 
25   trade-through rule, as it stands, provides anyone with a 
0056
 1   protected quote with pricing power.  And we see that 
 2   expressed through the fact that we need an access fee 
 3   cap, or it would likely continue to go up. 
 4             If thinking in terms of a trade-at provision, 
 5   that pricing power would become much more intense.  So, I 
 6   view that doing a pilot to assess the impact of the fee 
 7   cap as being a good prerequisite to even considering a 
 8   trade-at provision because it will give us an untainted 
 9   view -- or a less tainted view given that there is 
10   already a trade-through rule of the effective -- you 
11   know, of the fees on order routing or market quality. 
12             To Commissioner Stein's point and, I guess, 
13   contrary to some other opinions, this committee was 
14   looking at market quality for investors across retail and 
15   institutions.  And these recommendations come from that 
16   perspective, not the perspective of anyone who's self-
17   interested.   
18             So, I appreciate the opportunity to speak. 
19             MR. LUPARELLO:  Thank you, Adam. 
20             Not everybody at once. 
21             MR. CRONIN:  Can I just give --  
22             MR. LUPARELLO:  I'll let the subcommittee chair 
23   go first.   
24             Kevin. 
25             MR. CRONIN:  So, one thing I wanted to make 
0057
 1   clear -- so, when I talked about being inclusive earlier, 
 2   we really were trying to be as inclusive as possible, and 
 3   I know there's been some criticism around the lack of 
 4   inclusion of listed companies, not the easiest group of 
 5   people to solve for, as single person showing up on a 
 6   committee.  But we did try.   
 7             Actually, we had a gentleman who has very good 
 8   corporate experience who was part of the conversations, 
 9   but he came to us a couple of conversations in and said, 
10   "You know what?  I'm probably not the right person for 
11   this."  So, we have definitely tried, as we've gone 
12   through these proceedings, to make sure that there is 
13   some representation for listed companies. 
14             I might add that Met and I are both part of 
15   publicly traded companies.  With our Power Shares brand, 
16   we list a number of different things on exchanges.  Our 
17   fund managers routinely meet with the companies that they 
18   own substantial percentages of, so we have a pretty good 
19   read on the corporate side of the equation.   
20             So, I think it would be disingenuous to suggest 
21   that there isn't some level of representation both 
22   currently on the committee and certainly that we haven't 
23   has some aspirational interest to make sure that that's 
24   as specific as we can make it. 
25             So, all that said, I just have a question.  
0058
 1   There's been a lot of ongoing discussion here about the 
 2   level of participation we've had and, again, the level of 
 3   inclusiveness.   
 4             And I just have to ask Jeff, if you guys are so 
 5   interested in representing the best interest of clients 
 6   and investors of all sort of types, issuers, why don't 
 7   you show up to the conversations? 
 8             MR. DAVIS:  We appreciate the invitations.  But 
 9   we had wanted and asked to be full members of the 
10   committee and have full representation and not be 
11   selectively included or excluded.  We don't support the 
12   process of meeting in the dark and think that that is not 
13   healthy for the process or for investors.  And so, we 
14   were holding out for a better process. 
15             MR. LUPARELLO:  I will just add that I promise 
16   that all subcommittee meetings are perfectly well-lit.  
17   And maybe we'll sort of bring the process conversation to 
18   an end, although I appreciate, Kevin, your points -- and 
19   maybe get to the substance of the recommendations. 
20             Jamil then Rick. 
21             MR. NAZARALI:  So, getting our point of view on 
22   the access fee pilot, you know, I think we have some 
23   concerns about it, to echo what Adam and Tom Farley and 
24   Charles Jones said.   
25             Market makers don't view a dollar from rebates 
0059
 1   any differently than a dollar from trading profits.  And 
 2   so, when you are placing liquidity, you take into account 
 3   the holistic amount that you're going to get.  And so, we 
 4   expect that spreads will widen commensurately with the 
 5   reduction in the ability to pay a rebate.  Now, is that a 
 6   good thing or a bad thing?  Well, it depends on who you 
 7   are. 
 8             Some of this may result in wider spreads for 
 9   retail investors.  Some of it may be competed away so 
10   they get price improvement.  We just don't know. 
11             You're also going to see that retail broker-
12   dealers who place a lot of limit orders and get revenue 
13   from that that they then use to subsidize some of the 
14   products that they give to retail investors are going to 
15   get much less revenue.  So, that's another concern. 
16             Having said that, I'd like to address Chair 
17   White's questions on the actual pilot and the elements of 
18   the pilot.  
19             On the first question, should there be a bucket 
20   with no access fee, I think that that's actually really 
21   hard to do because you're saying to the exchanges, 
22   "You're not allowed to charge any revenue for this."  And 
23   if it went beyond a pilot and you said, "Okay, that's 
24   great.  It encourages a lot of liquidity," it's not 
25   something that you could implement.   
0060
 1             One thing you could consider is, you could say, 
 2   "No rebates are allowed to be paid," right.  So, that's 
 3   another way around that.  You could say -- you know, you 
 4   could have a bucket where a taker or a maker rebate 
 5   couldn't be paid. 
 6             The other thing about the no-access fee, 
 7   particularly if it's applied to other venues like ATSs, 
 8   that's really more of a commission, and I think that that 
 9   is going to be hard in practice because, if you limit 
10   access fees outside of exchanges, you then say to an ATS, 
11   "Okay.  Access fee is five or zero," how do you really 
12   regulate that, because many algo providers don't charge 
13   internally for their ATS.  They charge a commission, and 
14   they say, "Okay.  I'm going to charge you a quarter cent. 
15    I'm going to cross some in my ATS," but how much of that 
16   related to the ATS?    
17             And so, I think that, in practice, that's going 
18   to be really hard.  If you did regulate that, then do you 
19   regulate it also just for electronic trading, or do you 
20   say high-touch trading?  Because if you did it just for 
21   electronic, then that would then allow many firms to just 
22   charge that same fee for high-touch trading.  So, I could 
23   trade these pilot securities with the five-cent cap, but 
24   I'll do it high-touch.  So, I think that the -- being 
25   able to do that is really hard. 
0061
 1             One thing I think also is important to 
 2   consider, and that is, you know, Charles Jones spoke 
 3   about the market failure of conflicts of interest that 
 4   this addresses.  I think that it partly addresses that, 
 5   but you're still going to, in any world, have big 
 6   differences in fees, and that conflict of interest is not 
 7   going to go away.  And there's actually easier ways to 
 8   directly address that conflict of interest. 
 9             For example, if we required institutions to 
10   have cost-plus, that conflict would go away overnight.  
11   And yes, there are some complexities to doing that 
12   because exchanges have volume tiers, etcetera, but that's 
13   actually a fairly easy problem to solve, and certainly, 
14   we have algorithms that we chart cost-plus. 
15             So, it's a little bit complicated, but it's 
16   much easier to address.  And that actually completely 
17   addresses conflict of interest rather than partially 
18   addresses that, so that's something to consider. 
19             On Chair White's question of should we include 
20   a trade-at pilot, we don't think so because there's 
21   already a trade-at pilot in the tick test pilot.  And I 
22   think seeing how that goes is actually quite important. 
23             I think also, during the comment period for the 
24   tick test pilot, you got a lot of industry comment on 
25   what the industry thinks about trade-at.  And while we're 
0062
 1   strongly supportive of visible liquidity; it's critical 
 2   for what -- everything we do, we're really concerned that 
 3   trade-at limits competition and would result in less 
 4   efficiency. 
 5             It would also result in a huge transfer of 
 6   wealth from retail investors who right now get much 
 7   better prices than the lit venues because their orders 
 8   tend to be small and the orders that you get are in their 
 9   entirety, and so, there's an ecosystem that's evolved to 
10   give them better prices.   
11             If you had trade-at and you forced everyone to 
12   get the same price, then retail investors would 
13   necessarily get worse prices and other investors would 
14   get better prices.  So, that would result in a big 
15   transfer of wealth, and so, we're not really supportive 
16   of the trade-at.   
17             On the last point of when we should do this 
18   pilot, this pilot is really interesting, but one thing 
19   that we've heard a lot about is that, if you take away 
20   the incentive to provide liquidity, you're going to have 
21   less liquidity provision.  And so, we expect that that's 
22   going to be part of what happens.  Now, if you do that 
23   for the most liquid securities, it's probably not a 
24   problem because there's a lot of liquidity there.   
25             So, one really interesting part about this 
0063
 1   pilot would be to see what the impact would be on the 
 2   less liquid securities.  So, focusing this pilot on 
 3   greater than three billion market cap strikes me as 
 4   missing a really important component of what could 
 5   happen. 
 6             So, imagine this thought experiment, that the 
 7   pilot goes well and you say, "Okay.  Well, this is great. 
 8    I'm going to roll it out to the entire universe."  You 
 9   can't really do that because you don't know the impact on 
10   the smaller cap.  And that could be really, really 
11   important, because as we've all noted, there tends to be 
12   much less liquidity than all of us would like in these 
13   small caps.  
14             And so, what strikes us as a better way to do 
15   that is to include the small caps and to run this once 
16   the tick test pilot is done.  So, again, that's our 
17   overall view of that.   
18             Thank you. 
19             MR. LUPARELLO:  Rick? 
20             MR. KETCHUM:  Yeah.  Just a question for Kevin 
21   and Joe and other members of the subcommittee. 
22             In your decisions that you describe well here 
23   with respect to not including ATS access fees and not 
24   addressing taker-maker fee environments, the primary 
25   point articulated in the outline seems to be concern of 
0064
 1   the fact that it's not presently applicable with respect 
 2   to Reg NMS.  But given the breadth of the Commission's 
 3   authority under 11(a), if the Commission did determine 
 4   that it was appropriate to move forward, are there 
 5   substantive reasons that the subcommittee had to not 
 6   include either of those areas from the standpoint of an 
 7   access pilot? 
 8             MR. MECANE:  It's a fair question.  We 
 9   definitely spent a lot of time talking about it.  I think 
10   -- aside from just the authoritative reasons, I think the 
11   biggest reason we didn't include them was trying to be as 
12   true to being able to compare the results of the 
13   experiment versus how the market works today.   
14             And by changing a rule on top of the access fee 
15   that is currently not in effect, I think we were mainly 
16   concerned about the implementation details that that 
17   would entail and then how do you compare what the end 
18   result of the access fee experiment ends up being versus 
19   how the market works today.  So, it was more about the -- 
20   I would say the effectiveness of the pilot itself as 
21   opposed to just the rule reasons. 
22             If I can make one other point while I have the 
23   microphone, just to maybe address a few of the things 
24   that I think especially came from the panelists, because 
25   I think they're all valid points, but it just 
0065
 1   illustrates, I think, one thing which we, in our 
 2   conversations, ended up concluding, which is, where 
 3   everyone talks about, you know, access fees impacting 
 4   particular constituents of the marketplace, I think 
 5   they're all valid, right.   
 6             Market makers might make more money because 
 7   spreads widen.  Market makers might make less money 
 8   because rebates shrink.  ATSs may internalize less 
 9   because they are trying to avoid fees in a lot of cases, 
10   or ATSs may internalize more because, you know, their 
11   value proposition is different than economics. 
12             The point is -- and I think Adam said this, 
13   there are all these counterbalancing factors which we 
14   just don't understand.  And even, like, putting my 
15   Barclays hat on, access fees, to us -- we're roughly a 
16   50/50 net taker -- it doesn't really have an economic 
17   impact on us.  And I think a lot of people with their 
18   diverse business models -- it's not clear what impact 
19   access fees are going to have on their models. 
20             I would argue that, you know, even exchanges, 
21   retail, every one of the constituents of the marketplace 
22   are going to have certain factors affect them positively 
23   and certain factors affect them negatively.  The reality 
24   is, we don't know what the net impact of those are going 
25   to be.   
0066
 1             So, I think we all -- not to speak for 
 2   everyone, but I think we generally went into the 
 3   conversation with the perspective that we don't -- we can 
 4   draw certain conclusions about what behaviorally is going 
 5   to change.  But what then the impact of all of those are 
 6   going to be is very, very unclear. 
 7             So, I think, again, as I think Adam said, 
 8   gathering the data and understanding what is such a 
 9   significant part of our marketplace -- and we really 
10   don't understand well -- we think is the primary 
11   objective of the pilot. 
12             MR. STONE:  I just want to add one other thing.  
13             So, Commissioner Stein and Rick, you both asked 
14   a good question, which was timeline.  One of the reasons 
15   why we wanted to stick to what NMS was now was because we 
16   thought, right or wrong, it would be something that could 
17   potentially be done as an NMS plan.  When you start 
18   changing capping and rebates, which NMS doesn't address, 
19   we figured that would be a rule change which then would 
20   lead to a longer timeline. 
21             If we are incorrect in that assumption, then 
22   let us know, but that's what we generally got in our 
23   discussions with the staff and our discussions with each 
24   other, because we wanted to get something out there. 
25             MR. LUPARELLO:  And I suppose I could spend a 
0067
 1   minute talking about what a timeline would look like.  As 
 2   we've talked about, the subcommittee is making a 
 3   recommendation to the full committee.  Assuming the full 
 4   committee were to view that positively, that would, in 
 5   turn, become a recommendation from the committee to the 
 6   staff.   
 7             That's obviously something the staff would take 
 8   very seriously and consider in terms of developing a 
 9   proposal, but it would be a proposal like anything else 
10   that goes through the Commission process, which includes 
11   an analysis by division staff working in -- I was going 
12   to say working in partnership with Mark, but Mark's not 
13   sitting down there any more -- in terms of making a 
14   recommendation to the full Commission, which would then 
15   obviously have to go out for a separate round of 
16   comments. 
17             So, speed is very important, but it's basically 
18   speed to the starting line as opposed to speed to the 
19   finish line. 
20             Brad. 
21             MR. KATSUYAMA:  Sure.  
22             So, I think the lead-in by Kevin and Joe was 
23   actually pretty good.  Understanding that there are a 
24   variety of different interests, trying to find a way to 
25   balance those and also take a step forward isn't easy.  I 
0068
 1   think the recommendation is to keep things simple, i.e., 
 2   trying to change as few variables as possible is 
 3   sensible.  You know, conflating trade-at on top of this  
 4   -- so, again, it's going to -- what it does is it leaves 
 5   holes for people to question the results if results are 
 6   conclusive.  So, I think, again, it's very, I think, 
 7   sensible, well laid out.   
 8             I think there is possible consideration for a 
 9   zero rebate, part of this, only because it let's the 
10   market set the price as opposed to picking an arbitrary  
11   -- what is the right limit?  I think that's a question 
12   that most people ask themselves.  And I don't really know 
13   if there's a good answer to that. 
14             You know, the other part about liquidity in 
15   general -- I think that the term's used very openly.  
16   It's thrown around a lot.  You know, if you think about 
17   the times when liquidity is needed in times of duress, I 
18   don't know if you can pay someone to stand in front of a 
19   freight train.  I don't know if an extra 10 mils is 
20   really going to make a huge difference.  And, you know, 
21   since we don't have sub-penny quoting, changing the 
22   rebate, will it really necessarily change the ticks?  
23   Will it change it dramatically?   
24             I think it was sensible, again, to pick three 
25   billion and over.  I think, if the tick pilot wasn't on 
0069
 1   the agenda, I don't think that they would have picked 
 2   that threshold, but they're trying to be sensible about 
 3   timing, and maybe if you get something conclusive out of 
 4   this particular pilot, then you can look to change access 
 5   fees for three billion and above and run a pilot for less 
 6   -- you know, for stocks below that. 
 7             So, I think what they've laid out is a 
 8   reasonable compromise.  I think that the speculation that 
 9   spreads will automatically widen, especially with, you 
10   know, one penny increments.  I think probably 
11   underestimates the competition amongst market makers to 
12   post best price.  I think that, you know, again -- and we 
13   can speculate all we want, but I think this is -- this 
14   comes at a time where this will prove it.  I think it's 
15   pretty sensible. 
16             MR. LUPARELLO:  Thank you, Brad. 
17             Matt? 
18             MR. ANDRESEN:  All right.  So, first I just 
19   want to raise one concern I have with the concept of a 
20   zero-tier bucket.  And my concern is that, as Jamil 
21   noted, when you cap the exchange revenue at zero -- 
22   exchanges are self-interested entities like any other.  
23   They're going to look for another way to make that money. 
24     
25             And a concern I've raised on this committee 
0070
 1   before is the exchanges' propensity for mining non-
 2   transactional fees to the heart's content.  And while we 
 3   have focused and the Commission has focused on these 
 4   explicit fees on a per transaction basis, it's the 
 5   somewhat more hidden fees that provide a real barrier to 
 6   entry from new participants who could provide new price 
 7   competition. 
 8             So, things like the cost of ordering the SIP, 
 9   both SIPs, while they've been questioned greatly over the 
10   last couple of years, have also, at the same time, had 
11   the fees required of them go up dramatically. 
12             The cost of colocating in a data center has 
13   gone up dramatically.  The cost of cross-connects, of 
14   telco and everything has gone up.  So, to the extent that 
15   a well regulated, well understood revenue piece of the 
16   access fees goes away, my concern is that that might get 
17   pushed, to use an example, into the dark and might be 
18   even harder and more of a Gordian knot to have to 
19   untangle. 
20             One other -- I just had one question for one of 
21   our panelists just as a thought exercise.  I am -- on a 
22   personal note, I do find the NYSE's views on the 
23   potential value of a maker rebate refreshing.  Having 
24   dealt often with the prior management, it is nice to hear 
25   that there is a recognition of the potential relationship 
0071
 1   between a rebate and the quality of that market. 
 2             My question though to Mr. Farley is, if that is 
 3   a recognized market microstructure tenet at NYSE Group 
 4   and therefore within ICE, why not do that in the other 
 5   markets, like, say, agricultural futures?  Why is it only 
 6   good for equities?  Why not do it on the futures side? 
 7             MR. FARLEY:  The short answer is because it's 
 8   suboptimal.  In my career I have run futures exchanges 
 9   and credit default swaps, marketplaces, bond marketplaces 
10   and now equities and equity options.  And these are the 
11   only ones where you have this maker-taker construct, and 
12   it is suboptimal. 
13             If I could wave a magic wand or we could wave a 
14   magic wand at the NYSE and we could divine what market 
15   structure would be in a way that we think would be most 
16   advantageous to listed companies, we would actually 
17   eliminate rebates as a matter of course entirely, which 
18   is to say we would say, to the extent there's a fee -- 
19   you know, let's say it's one mil a side, the one mil a 
20   side would apply to both the maker and the taker, number 
21   one. 
22             Number two, if you wanted to incent market 
23   makers to make prices, you would file with the SEC 
24   exactly what the program you were proposing.  It would 
25   have real meaningful obligations.  In other words, on 
0072
 1   August 24th, those market makers can't disappear, which 
 2   they can do in a maker-taker construct where everybody 
 3   gets the rebate today and they have no meaningful 
 4   obligations. 
 5             And number three, you would motivate those 
 6   market makers even further to make those prices by 
 7   providing that, if they're the best price in the market, 
 8   somebody can't go and trade away from the exchange with 
 9   somebody else at their price, using their price as a warm 
10   blanket to make them comfortable in making that price.   
11             And that would be a much more optimal structure 
12   that I think our listed companies would really get 
13   behind.   
14             MR. LUPARELLO:  Sorry.  Adam first, then Jeff. 
15             MR. NUNES:  Great.   
16             If I can just address that, I don't presuppose 
17   an outcome to what happens in the pilot.  And I think 
18   that one of the benefits of the pilot could be 
19   demonstrating that value.  So, if we get the statistics 
20   at the end of the pilot and it demonstrates that, you 
21   know, 30 is the best, then that's where it should stay.  
22             And I don't feel like anyone should be 
23   concerned about putting it to that test to say, "Hey, 
24   maybe 20's better; maybe 10's better; maybe something 
25   lower than that's better."  We'll see what the data 
0073
 1   shows, and we can make a decision based on that. 
 2             MR. DAVIS:  I think it's important to also -- 
 3   Steve, to correct the record.  If what you want to talk 
 4   about is hidden fees, you would not be talking about 
 5   exchanges or SIPs, all right, because the exchanges and 
 6   the SIPs are required to file all their fees with the 
 7   SEC, which is why we're talking about access fees here in 
 8   the first place, because every fee that the exchanges 
 9   charge on any of our functionality is filed with the 
10   Commission, and so, everybody can see what it is. 
11             If what you wanted to do was to test hidden 
12   incentives or hidden conflicts or the agency problem, as 
13   the professor called it, you would look at a much broader 
14   sample.  You would look at algo fees.  You would look at 
15   ATS fees.  You would look at connectivity fees for ATSs. 
16    You would look at payment for order flow.  You would -- 
17   if what you want to do is look at the ecosystem of fees 
18   and see how it affects trading behavior, look at the 
19   ecosystem.  Don't just pick one limited variable that 
20   doesn't affect your own venue. 
21             MR. LUPARELLO:  Reggie then Adam. 
22             MR. BROWNE:  I have two comments.  My first 
23   comment, seems like this study is focused on corporate 
24   stocks.  And so, my comment around the study around the 
25   behavior of corporate stocks and intended consequences, I 
0074
 1   hope that we can also measure competition.  With the 
 2   rebate structure now, we do have competition in the 
 3   marketplace from retail broker-dealers who have used 
 4   rebates to enhance their business platforms.  I think 
 5   invariably there will be reduced competition or less 
 6   innovation if this study goes forward.  And that should 
 7   be a consideration.   
 8             Secondly, in terms of ETFs, while this study 
 9   does not address ETFs, thankfully, my concern is lower 
10   participation by providers offering liquidity.   
11             Ten percent, roughly, of the ETF are over three 
12   billion.  Another 15 percent are over a billion in market 
13   capitalization.  And when you talk about equity market 
14   structure, you have to also think about the ETF 
15   micromarket structure and changes to the rebate structure 
16   that would take place.   
17             And to highlight this effect, if you think 
18   about it, how ETFs are IPO'd in terms of a corporate 
19   stock, it's by market makers and APs.  Those market maker 
20   APs provided nearly a billion dollars of seed capital to 
21   the industry to propagate innovation and do ETFs last 
22   year. 
23             If you reduced the incentive for liquidity 
24   division in ETFs, you have lower opportunity to increase 
25   innovation, and also you will have wider spreads in the 
0075
 1   ETF market. 
 2             So, the study should also take into 
 3   consideration, especially if you're talking about 
 4   implementing two years from now, those consequences 
 5   around the ETF market. 
 6             MR. NUNES:  So, just adding to Matt's point on 
 7   the other fees, you know, the difference between an ATS 
 8   and an exchange is exchanges have protected quotes, which 
 9   has the effect for significant market participants.  You 
10   have to connect.  You know, we certainly connect to 
11   exchanges where, between the port fees, the trading 
12   rights fee, the membership fee, getting a data feed, the 
13   cost of connecting and the fact that they have a 
14   protected quote, is a few thousand dollars and the cost 
15   of actually trading there ends up being effectively de 
16   minimis in that fee conversation. 
17             So, you know, they get a toll both from Reg 
18   NMS, and the access fee isn't the only way they use it, 
19   and it's not going to be the only way they continue to 
20   use it. 
21             MR. LUPARELLO:  Joe and then Jeff. 
22             MR. RATTERMAN:  Thanks, Steve. 
23             So I want to -- a couple of comments.  First of 
24   all, on the simplicity matter, which has come up a few 
25   times, I'm very much in favor of keeping what we're going 
0076
 1   to do here simple.  The access fee was put in place with 
 2   Reg NMS based on the fact that certain exchanges have 
 3   protective quotes.  And it's been said before today that, 
 4   without some kind of regulatory intervention, those 
 5   venues that have a protected quote and an obligation for 
 6   everyone to come access their quote could charge whatever 
 7   ransoms and rents that they want.  And therefore 
 8   regulatory intervention makes a lot of sense. 
 9             I worry, as we start talking about adding 
10   trade-at and regulating rebates, maybe the size of the 
11   rebates or no rebates, that now the SEC is starting to 
12   get into market micropricing regulation.   
13             And I -- you know, I'm worried that that's not 
14   a place that the SEC or the regulator wants to be, that  
15   -- regulating where free competition could get our of 
16   whack, I think is very wise, and also it provides the 
17   simplicity to say, "We've regulated around this number 
18   before.  We can move it down.  It is an easy way to make 
19   this pilot happen." 
20             So, again, you know, if you start going down 
21   the path of, well, we're going to regulate all the 
22   elements of the way broker-dealers and exchanges charge 
23   the fees, that's a very long path, much longer than 
24   what's being contemplated here today.   
25             So, stopping short at saying, "Well, let's talk 
0077
 1   about rebates," I think, you know, wouldn't be far enough 
 2   if you're going to regulate everything about the way 
 3   exchanges charge. 
 4             So, I like the idea of keeping it simple and 
 5   focusing on the one fee that is there for a purpose, and 
 6   that is to keep venues from gouging participants over 
 7   obligation if they have to come and access their quote. 
 8             Jamil and I sat down with the retail folks 
 9   recently, and Faris espoused many of these same views.  
10   But I just want to make sure that we've been able to 
11   carry those messages forward. 
12             We had a good, long conversation about the 
13   topics today because they didn't have direct 
14   representation.  And five points out of several I'd like 
15   to just bring up that I heard loud and clear -- and Faris 
16   will have said a few of these today already. 
17             First of all, rebates and -- I'm sorry, spreads 
18   and commissions are great from the retail perspective.  
19   Nobody is complaining.  The retail experience is 
20   fantastic.  Their commissions are low.  Their spreads are 
21   tight.  Their experience is fantastic.  So, from the 
22   retail perspective, life seems pretty good with today's 
23   market structure. 
24             Secondly, from those that are on the ground 
25   that would have implement any things that come out, 
0078
 1   simultaneous pilots can be a real drag on the 
 2   organizations.  And so, sequencing pilots to test, you 
 3   know, a few things at a time, you know, is best.  
 4   Overlaying the multiple pilots could be a drain on 
 5   resources that are otherwise spent trying to address 
 6   their customers. 
 7             Defining success in advance is vital.  You 
 8   know, we all have, as I think others have said, views on 
 9   what might happen.  And I have my own views.  Adam has 
10   his views.  Some of them are the same.  We all have 
11   different views, but the fact is, is we don't know for 
12   sure.  But what we can be for sure of is that we have 
13   identified in advance what success would be if we happen 
14   upon it so that we can see it and what failure would look 
15   like so we can pull the plug quickly if we see that as 
16   well. 
17             So, I think a lot of effort, before we start a 
18   pilot, has to go into defining very clearly so that all 
19   participants agree what success actually was or wasn't 
20   before we start, because there's a real tendency, I 
21   think, to say what we saw is what we wanted or what we 
22   didn't want after the fact if we didn't define it in 
23   advance. 
24             And then, finally from the retail folks, the 
25   final point that I heard loud and clear was that they 
0079
 1   feel that a pilot will neither address the complexity nor 
 2   the conflicts that are involved in our markets, that 
 3   reducing access fees or changing rebates doesn't change 
 4   the complexity of the market.  In fact, another pilot 
 5   does add complexity.   
 6             And the conflicts are still there.  I think 
 7   we've heard that already today, that conflicts are going 
 8   to exist so long as you don't have a central limit order 
 9   book and a single monopoly who can charge whatever they 
10   want and that's the only place to go in town.  While you 
11   have competition, there will always be conflicts in it.  
12   You can't regulate those conflicts away.  You can try to 
13   mitigate them.  You can try and look for them, identify 
14   them, but you can't take them away. 
15             And then, finally, you know, our view is that a 
16   pilot does make sense.  We were part of the subcommittee. 
17    Keeping it simple, defining success, and getting a broad 
18   representation -- a broad consensus from all participants 
19   on what we're doing and how would be the reason to go 
20   forward.   
21             And so long as there's significant division in 
22   the industry about what it should be and how it should be 
23   taken on, we would -- we would advocate to holding back 
24   until we can achieve that consensus from across the 
25   industry. 
0080
 1             MR. LUPARELLO:  Thank you, Joe. 
 2             Let me do Jeff and Tom and then Gary. 
 3             MR. DAVIS:  Oh, I didn't have anything. 
 4             MR. LUPARELLO:  Oh, I'm sorry.  Then, turn off 
 5   your mike. 
 6             Then Tom then Gary. 
 7             MR. FARLEY:  Thanks, Steve.   
 8             I suppose echoing what now three people have 
 9   said, the retail investor is in a good spot, and there's 
10   a Hippocratic oath element to this, that, you know, do no 
11   harm.  But that's distinct from saying that we can't make 
12   changes and that we can improve.   
13             And so, I want to go back to something Jamil 
14   mentioned, that if you had a trade-at, you would 
15   necessarily diminish the experience of the retail 
16   investor.  I don't believe that to be true.  One of the 
17   things -- it's a very tricky balance.  If all of the 
18   retail came back into the broader market today, the rest 
19   of the broader market would likely have a slightly better 
20   experience, and retail would have a slightly worse 
21   experience.   
22             And one who isn't retail could argue, "Well, 
23   yeah, but that's what's best for the market."  But that's 
24   pretty difficult to tell Mom and Pop, "Hey, you need to 
25   sacrifice for everyone else.  You're now going to get a 
0081
 1   worse experience." 
 2             And so, it's a tricky balance, but I would come 
 3   into it with, again, the Hippocratic oath, predisposition 
 4   of, you can't do harm to those retail executions.   
 5             But having a trade-at that's -- I'll call it a 
 6   smart trade-at -- you know, when I -- I've seen some 
 7   execution reports in terms of how well retail gets 
 8   filled.   
 9             And I don't have these committed to memory, but 
10   I actually remember your firm does a really good job, 
11   Jamil, in getting the retail investor a market 
12   improvement from what is the prevailing NMS price.  And I 
13   think you and I, privately over a beer, could agree, if 
14   you're not giving an investor an improvement, you 
15   probably don't deserve to trade with them. 
16             And so, having a trade-at for retail that is 
17   set at a small enough increment that you can provide 
18   retail a price improvement; they can continue to get as 
19   good or likely even better an experience than they get 
20   today, might be a decent approach. 
21             When I have that conversation, not here in this 
22   room  but privately, perhaps over a beer, people do say, 
23   "You know, that's right.  But we see it as a nose under 
24   the tent.  This is you just getting in there.  And 
25   eventually you want to move all this business onto 
0082
 1   exchanges."  And so, I'm just putting it out there to 
 2   say, "No, absolutely not."   
 3             What can we do that solves the Hippocratic oath 
 4   equation of doing no harm for the long term, for the 
 5   retail investor but improves the overall marketplace?  
 6   And I reject the answer that the answer is "no, we can't 
 7   do anything". 
 8             MR. LUPARELLO:  Gary. 
 9             MR. STONE:  So, if you could, our 
10   recommendation number five is actually that -- addresses 
11   what Joe and a little bit of what Tom saying, which was 
12   that one of the things we want to do is build in criteria 
13   to terminate the program should there be a significant 
14   deterioration in any measurement of criteria.   
15             So, I think, in addition to defining success, 
16   we should define what failure is in any of the buckets.  
17   And we didn't really explore that one or be prescriptive 
18   about it because we thought that was something which  
19   would come out in the public comment to tell us when is 
20   it going horribly awry. 
21             But the one thing that is implied in this that 
22   I do want to make sure that the Commission notes is that 
23   one of the things that we would like to see is actually 
24   periodic reports from the Commission as to what is going 
25   on with the pilot so -- and I'm just going to throw it 
0083
 1   out there.   
 2             So, this isn't the subcommittee, this is my  
 3   -- me talking.  I would say it -- you know, month 6, 12, 
 4   18, and then at the end, 24 -- use six months' data 
 5   report; in month seven, what happened over the first six 
 6   months and you do that all along, that gives people an 
 7   idea where we are and where we need to pull the trigger 
 8   if it's not working. 
 9             MR. LUPARELLO:  And Mark walked in just at the 
10   right time for that ask.   
11             Met, did you have anything? 
12             Oh, sorry.  Kevin. 
13             MR. CRONIN:  So, just a couple of things.  As 
14   we're defining success, it's probably also helpful to 
15   define who we're trying to be successful for.  And we 
16   talk about long-term investors, but then in the next 
17   sentence, I always seem to hear "retail".  I think the 
18   record would probably accurately reflect that most of the 
19   investors that we're quantifying as retail get their 
20   exposure by using institutional products, right.   
21             So, this notion that somehow holding harmless 
22   is the sort of standard that we're trying to get to, 
23   we're standing here as institutions with our arms up, 
24   saying the current mechanism isn't optimal for our kind 
25   of participation.  And we are willing to take the risk 
0084
 1   of, in this instance, putting a pilot forth that may or 
 2   may not work.  It might cause some ripples in market 
 3   structure.  At this juncture, certainly we'd be willing 
 4   to take some of that risk to try to get to a better 
 5   place. 
 6             Our sense is that there are conflicts in the 
 7   marketplace -- they've been talked about already today -- 
 8   that are not being properly addressed.  Punting for two 
 9   more years isn't going to address them.   
10             So, we thought the best thing that we could do 
11   from our perspective -- and Met, I'll certainly let you 
12   speak for your own -- was to try to propagate something 
13   that would give us a break in the status quo here, 
14   something that was sensible, that was thoughtful, that's 
15   really designed to stop all of us theorizing about what 
16   this will mean and what the impact will be and be very 
17   scientific about it.   
18             We won't have to wax philosophical anymore on 
19   whether or not spreads will widen dramatically or whether 
20   or not Jamil's business model works.  We will be able to 
21   see very specifically, by the proposition of this and 
22   hopefully through the successful implementation of, how 
23   the access fee impacts all the various things that we 
24   believe need to be looked at at the very least. 
25             So, I would just close by saying this is an 
0085
 1   experiment that we are willingly taking on, recognizing 
 2   that it's not going to be perfect, recognizing that there 
 3   will be certain elements of it that will be 
 4   unanticipated.  But it's worth the risk to us, because we 
 5   believe, as long-term investors, there are elements of 
 6   the market that are just not optimal, and we need to try 
 7   to figure out how to optimize those things. 
 8             MR. LUPARELLO:  Met then Eric. 
 9             MR. KINAK:  Yeah.  I just want to opine on a 
10   couple of things, and so, while I have this -- I might go 
11   for a little while, Steve, just heads up.  Cut me off if 
12   I take too long. 
13             Let's start with --  
14             MR. LUPARELLO:  I'll read Eric's face, see how 
15   patient he's being. 
16             MR. KINAK:  Yeah, that's fine.   
17             Let's start with Mr. Farley's comments that 
18   maybe the subcommittee had some self-interested motives 
19   here.   
20             So, I think the Commission can tell you the buy 
21   side has been pressing this issue for, God knows, four or 
22   five years now that we've been in your doors with 
23   trillions of dollars under management, trying to get an 
24   access fee pilot.  And even in this case, Kevin and I 
25   were probably some of the biggest pushers when it came to 
0086
 1   this pilot.   
 2             So, to say that we're self-interested -- and I 
 3   think Kevin put it really, really well -- we don't really 
 4   have a self-interest in this.  There's a conflict in the 
 5   market.  I could guarantee you that.  I can show you that 
 6   in the data that I have, that there is a conflict in 
 7   routing that occurs in the market.  We are trying to get 
 8   rid of that conflict, right.   
 9             Does it work for us?  Sure.  It helps us manage 
10   our investments better.  But that's for the long-term 
11   investor.  That's for retail.  Let's be specific.  
12   There's retail and what we do.  And that's also for the 
13   benefit of firms like yourself.  Like, you want to talk 
14   about self-interested, we're the largest owners of ICE.   
15             So, to say that we have our own agenda and to 
16   say that we are only interested in what benefits our own 
17   investment, that's probably inaccurate, so I want to make 
18   sure we're clear on that. 
19             Secondly, so just going back to what this pilot 
20   was supposed to be, it's simple, right.  Take a bunch of 
21   stocks that are liquid that we already know that the 
22   spreads on those are artificially narrow, right.  So, we 
23   have a penny spread environment because we don't allow 
24   for sub-penny.  But we all know that that's not accurate. 
25    Jamil can tell you that, and Adam can tell you that as 
0087
 1   well, that they are artificially narrowed by the rebate 
 2   that's out there. 
 3             If you were to just take the rebate away from 
 4   those names -- and if we don't want to go into changing 
 5   rules and we want to just reduce the access fee -- that 
 6   makes things a lot simpler for us from a routing 
 7   perspective because I know that people won't do stupid 
 8   things for one or two mils, hopefully.  And I do agree 
 9   with Joe that you can never take the conflict out 
10   completely, but you can minimize it.  And they need to 
11   minimize it to a point where it's not incenting people to 
12   do the wrong thing.  And that's the simplest pilot that 
13   we wanted to throw out there. 
14             Now, when we go back to a question that, I 
15   think, Rick asked earlier, why didn't we include 
16   inverteds; why don't we want to include darks, everything 
17   should be included.   
18             The reason that other venues don't get included 
19   in this guise of "we have to change Reg NMS, and it was 
20   written in 2005," and blah, blah, blah, honestly, it just 
21   has to be a straight, narrow, every exchange, every 
22   platform, everything has to be equal, period.   
23             The reason that has to be is because, to Joe's 
24   point, you can't get rid of the conflict if you have 
25   certain venues that can do certain things that are 
0088
 1   differentiated from other venues.   
 2             So, if the inverteds can allow themselves to 
 3   keep providing rebates and providing fees at a certain 
 4   level that are different than what we're mandating with 
 5   bucket four, bucket two, or whatever it might be, that 
 6   changes, again, the order routing sequence.  That's all 
 7   we're really doing.   
 8             That's what you saw with the NASDAQ pilot.  
 9   They took themselves from the very end of the order 
10   routing sequence and put themselves right behind the 
11   inverteds.  That was the only change, and that's why you 
12   didn't see a material difference in any kind of metrics 
13   that they studied. 
14             But what would be interesting is to do it 
15   across the board and see whether or not there's actual 
16   change in routing preferences.  And that's the point that 
17   everyone's made now.  We don't know.  We don't know if 
18   it's going to go more dark.  We assume spreads might 
19   widen.  Maybe they actually widen to where they should 
20   be, which is a penny.  Maybe they don't at all.  Who 
21   knows?  Maybe we do trade a lot less in lit.  Maybe we 
22   trade more in lit because there's less fee avoidance. 
23             And then, finally on the last point Jamil made 
24   on getting -- just going to cost-plus, so, as an 
25   institutional investor -- so, you don't trade for 
0089
 1   clients, right, you trade for yourself obviously.  It's a 
 2   lot easier when you're collecting all of these fees or 
 3   paying or whatever it might be.   
 4             It's a lot different when you have thousands of 
 5   clients that, at the end of the month now, a fee change 
 6   comes in and says, "Well, you know what?  You hit certain 
 7   tiers with this broker.  You didn't hit it with that 
 8   broker.  This broker is able to give you now X percent 
 9   dollars back, and the other broker is actually not.  By 
10   the way, apply that to your thousands of trades that you 
11   did over a thousand clients going back in."   
12             MR. NAZARALI:  That's solvable though. 
13             MR. KINAK:  Well, it's not --  
14             MR. NAZARALI:  That's completely solvable. 
15             MR. KINAK:  So, that might be solvable for a 
16   firm like T. Rowe.  As you go down the spectrum of other 
17   firms that all have this differentiated thing, it's a lot 
18   more solvable to say, "By the way, why don't we just 
19   reduce the access fee to two and -- or zero or whatever 
20   it is and see what happens there. 
21             MR. NAZARALI:  But that's not going to address 
22   the conflict.  It's solvable.  And if you solve it, all 
23   you have to -- you can make one easy change.  You say, 
24   "The tiers apply to the next month."  Make that change.  
25             MR. KINAK:  How is that possible for the 
0090
 1   clients that didn't trade that month? 
 2             MR. NAZARALI:  It doesn't matter.  I mean, 
 3   we're talking --  
 4             MR. KINAK:  Well, wait a minute.  So, I'm 
 5   differentiating on which clients get paid for the trades 
 6   of other clients that were placed a month before. 
 7             MR. NAZARALI:  The tiers don't make that much 
 8   of a difference.  They don't vary that much month to 
 9   month. 
10             MR. KINAK:  But the rebates that I might be 
11   receiving might be significant for certain clients versus 
12   other clients when you look at the spectrum of what's 
13   being traded. 
14             MR. NAZARALI:  I think it's very minor.  I 
15   think it's an easy problem to solve.  And if you put the 
16   mindset to solving it, you could solve it.  Work with -- 
17   Joe will work here with you in the -- and he'll figure 
18   out a cost-plus because we've done it for lot's of 
19   clients.  It's not that hard to solve. 
20             MR. KINAK:  You've done it for institutional 
21   clients with multiple clients that they trade for? 
22             MR. NAZARALI:  Yes.  And I'd be happy to talk 
23   to you about it. 
24             MR. KINAK:  And then, one final point.   
25             So, Tom, you mentioned, you know, retail gets 
0091
 1   an exemption just because potentially they get a benefit, 
 2   right, if Jamil's willing to price-improve them.  That's 
 3   basically the counterargument of why people trade in 
 4   dark, right, and that's why you don't want them to trade 
 5   in dark.   
 6             So, if they can go in and get a midpoint or, 
 7   you know, a interest spread trade, that might be to their 
 8   benefit than just taking the bid and the offer.  But in 
 9   that case, you would propose to have a trade-at because 
10   it disincentivizes someone from posting.  But in the 
11   retail side, it makes sense because they're getting price 
12   improvement.  Seems a little bit -- you know, kind of --  
13             MR. FARLEY:  Yeah, that's tricky.  That's one I 
14   would want to have a debate in the open.  That's exactly 
15   what I was referring to. 
16             You know, for you, it would probably be helpful 
17   if retail was entirely in the public market.  But making 
18   the argument to the actual retail investor is difficult. 
19             MR. LUPARELLO:  Eric and then Faris. 
20             MR. NOLL:  Met hit on some of the points that I 
21   was going to make, but I think one of the points that we 
22   sort of missed talking about in all of this -- we've 
23   talked about the conflict and fee avoidance.   
24             But there's actually -- with the rebate 
25   structure as it currently is designed, there are actually 
0092
 1   distortions in trading, and we haven't really looked at 
 2   that or really spoken about that.   
 3             So, I won't use any current stocks because I 
 4   don't want to offend anybody, but when you look at 
 5   Citigroup, when Citigroup was a $4 stock, the dollar 
 6   value -- not just the share count -- but the dollar value 
 7   that Citicorp traded because of the ratio of rebate to 
 8   value of the underlying security was so skewed that the 
 9   tremendous -- hundreds of millions of shares traded of 
10   that stock every day.  The minute it reverse-split and 
11   went to a $40 stock, that disappeared.   
12             And so, what that highlights for me as I look 
13   at this and I look at the rebate structure is that there 
14   is some level of rebate where you are incenting 
15   distortions in trading.  And I think Kevin's pilot and 
16   the subcommittee's pilot is really designed -- is -- 
17   where is that breakpoint?  Where are you measuring those 
18   distortions that -- and Citicorp was the most obvious 
19   one.  But there are clearly others, and clearly that 
20   impact ripples through the system. 
21             And so, the idea here, at least from my 
22   perspective in supporting this pilot, is to look at how 
23   do we minimize that distortion, because I think that's 
24   going to be a key part in whether we're looking for real 
25   market quality here. 
0093
 1             MR. LUPARELLO:  Faris, maybe the last word? 
 2             MR. MATALKA:  I do want to comment on the 
 3   conflict of interest, and I want to make that distinction 
 4   between brokers that execute for buy side and retail 
 5   brokers.   
 6             Obviously, Reg NMS, Rule 611, really eliminated 
 7   a lot of these trade-throughs.  So, no matter where you 
 8   post an order on any exchange, the likelihood of that 
 9   execution, if it trades through, is almost 100 percent.  
10             I guess I think the -- you know, the Batalia 
11   paper about the -- you know, that edge case where if the 
12   price is touched throughout the day and prints are going 
13   off on exchange at that price, we do measure that.  We do 
14   measure that for our specific orders.   
15             And I want to cite some statistics here from 
16   January.  That exception ratio on three exchanges that we 
17   utilize is about five percent, is actually -- to be 
18   exact, NASDAQ 5.22, Bats 4.93, and Arca 6.06. 
19             So, you can see the distinction between -- I 
20   think it's an important stat -- to the difference between 
21   where you post that order -- is almost de minimis on the 
22   three exchanges that we utilize.  So, I think it's 
23   important to make that distinction when we talk about -- 
24   when we lump brokers -- I think retail brokers versus buy 
25   side brokers.   
0094
 1             So -- and then, to your point Tom, I just want 
 2   to make sure -- and Jamil as well, I think, while the 
 3   price improvement is pivotal for our business, it's not 
 4   the only factor, right.  There's factors that dealers can 
 5   commit capital.  There's the unquantifiable -- you know, 
 6   service desk, technology, the ability -- seamless to move 
 7   orders from one destination to another destination.  That 
 8   is also, in times of, you know, volatility and times of 
 9   technology instability that really factors into our 
10   decision to -- and why that model works today. 
11             MR. LUPARELLO:  So, let me wrap this up with a 
12   suggestion and maybe an ask of Kevin.  I think this has 
13   been an outstanding conversation.  I don't know that I'm 
14   hearing a groundswell of opposition to what the 
15   subcommittee has recommended.  The subcommittee has 
16   provided us with a great outline. 
17             If I could ask you to convert that outline into 
18   an actual recommendation, provide that to us -- and we 
19   can talk about this during the break.  But optimally, in 
20   the next week or two weeks, we can get that posted.  We 
21   can get back together either telephonically or in person 
22   to actually have a vote on that within a couple of weeks 
23   after that.  I think that might be a useful way forward. 
24             Is that something the subcommittee can do?  And 
25   obviously, if that's a -- that's either a time frame or 
0095
 1   an approach that's inconsistent with any of the views of 
 2   the specific committee members, I'd love to hear that. 
 3             MR. CRONIN:  Yeah, I think that works. 
 4             For New York and NASDAQ, I would offer again 
 5   for you to both be participants in this conversation.   
 6             MR. FARLEY:  Thanks, Kevin. 
 7             MR. LUPARELLO:  Thank you, Kevin.   
 8             I think with that we are at a lunch break.  So, 
 9   back here at 12:30.  Thanks very much. 
10             (Whereupon, at 11:30 a.m., a luncheon recess 
11   was taken.) 
12                A F T E R N O O N  S E S S I O N 
13             MR. LUPARELLO:  Welcome back.  We'll start 
14   again.  I do so with slight trepidation, looking at the 
15   empty seat to my left, but we'll go forward anyway. 
16             The afternoon session will follow the same 
17   format as this mornings, with an overview of the Trading 
18   Venues Subcommittee recommendations, panelist statements 
19   and then, I'm sure, a full committee discussion. 
20             So, with that, Rick, if I could turn it over to 
21   you. 
22             MR. KETCHUM:  Thanks Steve, and appreciate the 
23   attention the Commission's given to this committee.  
24   Appreciate all the efforts of my subcommittee members who 
25   participated consistently through each meeting, who knew 
0096
 1   -- until this morning, we did even know that plan rule 
 2   stuff was that controversial, but learning is everything. 
 3             A couple things up front, just to note the -- 
 4   where we sit in our considerations and clarify something 
 5   that was said this morning that may have been a 
 6   misunderstanding where sit, obviously, the committee was 
 7   formulated to look generally with respect to trading 
 8   venues with appreciation of not duplicating the work of 
 9   other subcommittees with respect to a variety of other 
10   issues that could implicate trading venues. 
11             It also -- like our subcommittee this morning, 
12   also tried to bring in representatives that represented 
13   both the industry, the ATS community and the exchange 
14   community over and above the members of the committee 
15   with respect to gaining input with regard to all our 
16   recommendations.   
17             We did, as well, invite NASDAQ and New York 
18   Stock Exchange and appreciate the reasons for their 
19   choosing not to participate but regret that -- and the 
20   effort had been to do that within the context of a 
21   presentation in which representatives of each of those 
22   areas would be there together, allowing them to respond 
23   to each other. 
24             We do appreciate that CBOE did participate in 
25   that, so we had the benefit of one large major exchange 
0097
 1   participating. 
 2             Also, just from a context standpoint, as the 
 3   document in front of you states, while the subcommittee 
 4   believed and supports the range of basic commission 
 5   decisions that have encouraged lower barriers to 
 6   accessing competition across both exchange venues and ATS 
 7   venues, it did recognize that there are a variety of 
 8   potential conflicts and issues and simply implementation 
 9   questions that are very important to look at with respect 
10   to the operation of venues. 
11             To clarify, the committee did not make a 
12   decision not to look at the appropriate regulatory 
13   environment for ATSs, although they did recognize and 
14   believe that the general structure built around ATSs and 
15   exchanges was appropriate.  Instead, the committee 
16   determined to defer further evaluation from the 
17   standpoint of ATS regulatory issues in light of the 
18   Commission's, we thought, very thoughtful proposed 
19   amendments attempting to address a variety of areas that 
20   had been identified through both commentary and 
21   Commission enforcement actions. 
22             And while the committee didn't take a position 
23   on the ATS amendments, it did believe it was going in a 
24   direction that was positive and wanted the opportunity to 
25   fully evaluate the comments that were received before 
0098
 1   making any recommendations with respect to further issues 
 2   from the standpoint of the regulatory structure for ATSs. 
 3    So, that will continue to be an item that the 
 4   subcommittee reviews in the coming months. 
 5             I also note that we look forward to gaining 
 6   access to a range of additional opinions from the 
 7   standpoint of each representative of the industry, the 
 8   investor community. 
 9             And while I miss the fascination of issuers 
10   over SRO plan issues in my 19-plus years and NASDAQ and 
11   on New York Stock Exchange, we would certainly welcome 
12   any suggestion of issuers that would love to talk about 
13   SIP oversight. 
14             To maybe take a brief moment on the 
15   recommendations that we provide, the first focus 
16   generally on the questions of self-regulatory 
17   organization immunity and the rule-based exchange 
18   liability limits that have existed and been adjusted over 
19   a period of years. 
20             The subcommittee at this point wasn't 
21   comfortable or able to reach a full consensus with 
22   respect to the immunity issue, so it had no views to 
23   express at this time. 
24             As actually the full consensus is probably 
25   worth a clarification, again, from some of the mis-
0099
 1   impressions that may have been suggested this morning, 
 2   each one of the subcommittee's recommendations was done 
 3   by not just a majority but a full consensus agreement 
 4   across all subcommittee members with one exception, and 
 5   that is the fourth item that I will pass on to Maureen 
 6   O'Hara because I chose to recuse myself, given FINRA's 
 7   particular interest in regulation that that 
 8   recommendation addressed. 
 9             So, each of these recommendations were agreed 
10   to by every member of the committee with respect to 
11   whatever self-interest or not-self-interest they brought 
12   into the conversation. 
13             With respect to exchange liability levels, we 
14   made a couple of recommendations that basically recognize 
15   that this environment has become more complex in an 
16   environment in which trading has increased and which 
17   active and high frequency trading has been substantially 
18   increased and algorithmic trading as well and that the 
19   questions and concerns from the standpoint of errors and 
20   decisions that an exchange can make just as the questions 
21   that are -- involve that from the broker-dealer or ATS 
22   broker-dealer side are substantial. 
23             And for that reason, we suggested that -- a 
24   recommendation that the liability level should be 
25   reviewed and increased to reflect this higher level of 
0100
 1   activity and also that they be more consciously applied 
 2   clearly and consistently across all self-regulatory 
 3   organizations, recognizing that that has not always been 
 4   essentially a requirement of entry from the standpoint of 
 5   an exchange in prior times. 
 6             While we suggested that those liability levels 
 7   should be consistent across all SROs, we did recognize 
 8   that some SROs that I will, for shorthand, refer to as 
 9   listing and auction market opening and reopening SROs 
10   are, because of the concentrated nature and the decision-
11   making around those type of activities, we suggested that 
12   those liability levels should be consistent with those 
13   exposures and should be reviewed separately and 
14   considered higher levels. 
15             We also recommended that, without suggesting a 
16   final approach, recognizing the complexity of it, that, 
17   just as the Commission continues to struggle with the 
18   challenges and the questions of appropriate financial 
19   responsibility levels for broker-dealers, that this is an 
20   area that should be reviewed with regard to exchange SROs 
21   with these type of liability limits and that, while that 
22   might not -- while exchange activity does not lend itself 
23   terribly easily to the type of net capital type of 
24   requirements the Commission had set before, that there 
25   should be a review and consideration as to whether some 
0101
 1   set-aside expectation should exist and with that an 
 2   expectation that, to the extent that there is an -- as is 
 3   the case with respect to some of these funds -- and I'll 
 4   note New York Stock Exchange does have some set aside -- 
 5   that money unused in one year would continue to be 
 6   retained and maintained separately on a continuous basis 
 7   thereafter, that that would also be considered by the 
 8   Commission.   
 9             Our second set of recommendations focused on 
10   the nitty-gritty of national market system planned 
11   governance.  And the recommendations generally focused on 
12   the structure and the role of the NMS Plan Advisory 
13   Committee that respects constituents and participants in 
14   the market who are not self-regulatory organizations and 
15   thus not, as a matter of the requirements of the plan 
16   rule, ever able to be members of the plan. 
17             And the effort with respect of each of these 
18   was to balance the appropriate responsibilities of the 
19   exchange and FINRA from the standpoint of SROs with 
20   regard to making the plan with the recognition that, as 
21   plan requirements and the range of plans have expanded to 
22   a range of extremely complex matters as well as continue 
23   to be important with respect to things like quote and 
24   trade reporting and the like, that the potential impacts 
25   on other constituents, particularly industry 
0102
 1   intermediaries but other constituents generally, are 
 2   substantial. 
 3             And therefore, identifying the appropriate way 
 4   for those industry representatives to be able to provide 
 5   input into the plan process, we believe, serves some 
 6   additional analysis. 
 7             That led to a suggestion, first, that the 
 8   Commission clarify the process for collecting, selecting 
 9   advisory committee representatives, that there be a 
10   clear, open process from the standpoint of advisory 
11   committee members or potential members indicating their 
12   interest prior to selection; that once the advisory 
13   committee was established, the subcommittee believed that 
14   the advisory committee itself should nominate its own 
15   replacement candidates.  Given the diversity that that 
16   committee had, we thought that that would be valuable.  
17             And that -- our final suggestion is that the 
18   selection should be approved by a simple majority of the 
19   Operating Committee. 
20             We also made a series of recommendations 
21   regarding expanding and formalizing the role of the 
22   advisory committee, again, not with any disrespect to the 
23   critical role that the SROs play in managing the 
24   decision-making process with regard to that but 
25   recognizing the impact of these plans on other 
0103
 1   constituents and participants in the market. 
 2             And I want to emphasize in saying these things, 
 3   none of this does not recognize that, with respect to a 
 4   variety of plan efforts, there has been an effort to 
 5   include advisory input substantially from the standpoint 
 6   of the committee and that this is not to suggest there 
 7   has been a conscious effort to try to exclude that impact 
 8   but only to regulize it and to create a structure around 
 9   it. 
10             So, we recommended that the advisory committee 
11   should have the right to a formal vote before any matter 
12   in which the operating committee votes; that if the 
13   operating committee subsequently approves any action that 
14   was opposed by a majority of the advisory committee, that 
15   the operating committee should explain and document its 
16   reasons for proceeding contrary to advisory committee 
17   input; and in the matter that the matter is a subject of 
18   a rule filing, that the operating committee should 
19   specifically summarize and explain the results of the 
20   committee and the reasons for its decisions that were 
21   different than the advisory committee goal. 
22             Now, again, this is not to suggest that this 
23   often is not done and that this is not part of the 
24   conversation between various individual plans in the 
25   commission but to regulize that. 
0104
 1             We also provided a range of recommendations to 
 2   significantly narrow the use of executive sessions by the 
 3   National Market System plan or, if you will, to have less 
 4   of that activity be done in the dark.  So, I'm quite 
 5   confident that this will be received enthusiastically by 
 6   all exchange members. 
 7             We first suggested that the advisory committee 
 8   members should -- generally should have the right to 
 9   attend all meetings and receive all information 
10   concerning plan matters distributed to participants in a 
11   time frame sufficient to allow them to evaluate them, I 
12   might add, except for executive sessions and other 
13   specified related materials that I'll address in a 
14   minute. 
15             We also suggested that the plan should limit 
16   the acceptable use of executive sessions, which has been 
17   a matter of some contention and lack of clarity with 
18   respect to the operation of the plans.   
19             And it should be limited to only matters that 
20   present a clear conflict for advisory committee members 
21   and where that conflict can be specifically established 
22   and described and only in those situations where a two-
23   thirds super majority vote of the operating committee 
24   concludes that those conflicts are real. 
25             We think that's valuable both from a procession 
0105
 1   and structural standpoint because we so strongly believe 
 2   that it's a better conversation if everybody participates 
 3   in it and so that we think there should be very, very 
 4   limited instances in which everyone doesn't have the 
 5   opportunity to participate in the conversation even 
 6   though the SROs have a unique responsibility as the 
 7   plan's operating committee.   
 8             We also did recognize though that there are 
 9   circumstances, somewhat like subcommittees, where there 
10   is value in creating ongoing working groups because, 
11   without them and without the ability to have informal 
12   conversations, it's sometimes difficult to -- either 
13   reach some level of consensus or, frankly, work through 
14   constructive ways to address different views across the 
15   group.   
16             So, we're suggesting that there should not be 
17   working groups.  We do suggest that working groups -- 
18   though it would not require advisory committee 
19   participation, that it must submit regulatory regular 
20   updates, sort of like subcommittees, and it's ultimate 
21   work product to the full operating committee and advisory 
22   committee so that the operating committee -- and finally, 
23   the operating committee must allow an advisory committee 
24   vote prior to taking any action on a working group 
25   recommendation. 
0106
 1             We have a separate recommendation to limit what 
 2   sort of has been more of a historical anomaly of pieces 
 3   of NMS plans which continue to require unanimous votes.  
 4   We believe that a two-thirds supermajority vote approach 
 5   would be more constructive, given the changes in the 
 6   environment that's occurred with respect to exchange 
 7   competition. 
 8             And finally, we address the sort of anomaly 
 9   that's resulted for perfectly appropriate reasons that 
10   exchanges have developed numerous registrations in order 
11   to provide flexibility from a competitive stand point to 
12   provide different pricing arrangements.  But that doesn't 
13   strike us as terribly logical from the standpoint of how 
14   the operating committee should operate from a voting 
15   standpoint.  And we are not interested in creating 
16   incentive to have nine million exchange registrations. 
17             So, we suggested that the present model should 
18   be replaced by a model in which there's an allocation of 
19   voting rights at the exchange group level, i.e., one for 
20   each exchange family but with a recognition that 
21   exchanges that are -- have a substantial amount of market 
22   share across their family should be recognized as having 
23   a particular interest from the standpoint of decisions 
24   that are made.   
25             And so, we suggest that -- without suggesting 
0107
 1   there's something magic to the 10 percent number, that 
 2   exchanges that accumulate across their registrations, 10 
 3   percent or more in the particular marketplace, should 
 4   receive an extra vote or, if you will, two votes instead 
 5   of one. 
 6             Our third recommendation recognizes the other 
 7   challenge with respect to plans and the decision-making 
 8   process here, that, as we have moved into a variety of 
 9   different responsibilities of plans, many of them 
10   generate substantial technology challenges for 
11   constituents in the market responding to these changes. 
12             Now, obviously, those constituents get the 
13   opportunity to express their views from the standpoint of 
14   notice and comment, but just as exchanges and FINRA 
15   provide opportunity for committees for more interaction 
16   than that, we think it's appropriate for there to be 
17   greater clarity on technology changes with the clear 
18   desire to have a greater opportunity for industry 
19   participants to respond and comment on technical 
20   specifications.  
21             And therefore, our recommendation suggested 
22   that draft technical specifications should be published 
23   prior to SEC approval of any related rule change and, 
24   where possible, that the industry members should be 
25   provided an opportunity to review and provide comment on 
0108
 1   those draft technical expectations.  
 2             And we also suggest that, except in limited 
 3   circumstances that necessitate expedited implementation, 
 4   that the implementation date of technology-driven rule 
 5   changes should be linked to the issuance of final tech 
 6   specs. 
 7             Now, obviously, that has to be done with some 
 8   balance and recognition of the importance of moving 
 9   forward, but we think that those basic rules should be in 
10   place. 
11             And our final suggestion is, where possible, 
12   the duration of the implementation period should be 
13   determined after the draft specs or related FAQs are 
14   issued to allow the industry to better evaluate and 
15   determine the necessary time frame. 
16             I'm going to pass our fourth recommendation, 
17   because it does implicate SRO regulatory oversight, to 
18   Maureen, noting that, while I look forward to 
19   participating in the full committee with respect to the 
20   views of an SRO regulator, I thought it was entirely not 
21   appropriate for the subcommittee's discussions for me to 
22   participate in it. 
23             So, Maureen? 
24             MR. LUPARELLO:  Maureen, microphone. 
25             MS. O'HARA:  Yep, that will help.   
0109
 1             Our fourth recommendation really thinks about 
 2   the question of what sorts of activities should be done 
 3   within the SRO structure and what should be done more 
 4   centrally.   
 5             And the -- I think there's a number of reasons 
 6   why this is an important topic today.  Certainly, one of 
 7   the issues is that, as exchanges have now become profit-
 8   seeking entities, the issue of whether there are 
 9   conflicts in which profit-seeking entities are also the 
10   regulatory entity is one that makes sense to address. 
11             And also, obviously, it is the case that a 
12   variety of activities at the SRO level have actually been 
13   sort of transferred over to FINRA in various way, shape, 
14   or form over the last few years.   
15             So, the committee really tried to think about 
16   where do you draw the lines.  And as will become 
17   apparent, it's not immediately obvious where all these 
18   lines are drawn.   
19             One of the things that we did discuss is, in 
20   light of the for-profit structure of exchanges, should 
21   you just simply say the SRO structure doesn't make sense 
22   anymore and go to a single regulator.  And our 
23   discussions here actually were in general agreement, that 
24   that's -- that wasn't, in our view, the right solution, 
25   that the SRO functions at the exchange level do play an 
0110
 1   important role with respect to a variety of activities. 
 2             One of them, for example, is real-time 
 3   surveillance activities that preserve and maintain the 
 4   fair and orderly market that the exchange wants to 
 5   maintain and obviously is something that all of us would 
 6   like to see occur. 
 7             There is also the example that the exchange has 
 8   strong interest in monitoring the compliance with listing 
 9   standards and other things to maintain, again, the 
10   quality level of the exchange.  So, our subcommittee was 
11   convinced that there are certain issues that do indeed 
12   remain for an SRO at the exchange level.   
13             But other things clearly do seem as if they 
14   might be better at a centralized level.  And certainly, 
15   some of those are apparent when you're going to have 
16   cross-market surveillance.  It's really difficult to see 
17   how that wouldn't be better done with a centralized 
18   focus. 
19             But there are a variety of other things that 
20   also seem to fall into this category.  And I believe that 
21   discussion over time will help delineate these even more. 
22             A variety of things that result in duplication 
23   of effort seem to us to be needlessly complex.  For 
24   example, order marketing, account type indicators, those 
25   sorts of things now are done individually by the 
0111
 1   different exchanges and that just seems needlessly 
 2   complex and duplicative.  We think that eventually, with 
 3   the rollout of CAT, this duplicative regulatory oversight 
 4   is going to be increasing.   
 5             And so, where we are right now in our thinking 
 6   is just sort of focusing on trying to think of things 
 7   that really do benefit from the centralization such as 
 8   the cross-market and that benefit from reducing 
 9   duplicative effort.   
10             But we are still -- as Rick mentioned, we are 
11   not entirely in agreement on where those lines would be 
12   drawn, and I think that more discussion, perhaps today, 
13   will help us get a better feel for that. 
14             MR. LUPARELLO:  Last words, Rick, or -- 
15             MR. KETCHUM:  No, that's it.  Thanks. 
16             MR. LUPARELLO:  Thank you, Maureen. 
17             Let me quickly introduce our panel.  This 
18   afternoon we have going, again, from left to right, David 
19   Cushing, who is the director of Trading and Market 
20   Strategies at Wellington Management.  To his left, a 
21   repeat performer, we have Jeff Davis, Deputy General 
22   Counsel of NASDAQ, and to -- next to him is Brett 
23   Redfearn, who is the head of Market Structure Strategy at 
24   J.P. Morgan Securities. 
25             So, David, can I ask you to start us off? 
0112
 1             MR. CUSHING:  So, Chair White, Commissioners, 
 2   Director Luparello, members of staff and members of the 
 3   committee, thank you for the opportunity to join today's 
 4   meeting and participate in this session on the subject of 
 5   improving current regulatory models and structure for 
 6   trading venues. 
 7             As Steve said, my name is Dave Cushing, and I 
 8   serve as director of Trading and Market Strategies at 
 9   Wellington Management Company.  Wellington Management is 
10   a global advisor, global asset manager, with assets under 
11   management of 943 billion as of March 31 of this year. 
12             Trading and Market Strategies Group at 
13   Wellington is responsible for trade order execution 
14   strategy and market structure engagement across asset 
15   classes globally.  Part of my role is to coordinate our 
16   firm's views and engagement with regulators, industry 
17   organizations, and other market participants on market-
18   related regulation, technology, and business practice. 
19             In my brief comments I will share our view on 
20   the subcommittee's overall assessment of whether the 
21   current regulatory model for trading venues is optimally 
22   serving the market and their specific recommendations.   
23             Let me start by thanking and commending them 
24   for their diligent work to address this complex subject, 
25   and let me also acknowledge the difficulty of recognizing 
0113
 1   such changes while being fair to all stakeholders and 
 2   remaining mindful of unintended consequences. 
 3             Our firm's advocacy and views are derived from 
 4   what we believe to be our clients' best interest.  As 
 5   fiduciaries, we are charged with acting in their best 
 6   interest, which impacts how we engage with markets, from 
 7   seeking best execution of orders placed for client 
 8   accounts to advocacy on the broadest market structural 
 9   and regulatory issues. 
10             First, on the general topic of SRO governance 
11   reform, we appreciate the difficulty of changing a system 
12   that is highly complex and ingrained in U.S. market 
13   structure.  However, the nature of exchanges has evolved 
14   so significantly since the current governance system was 
15   designed, that the SRO governance structure will clearly 
16   benefit from scrutiny and redesign in many areas. 
17             New conflicts have arisen as a result of this 
18   evolution, and these conflicts represent a significant 
19   challenge to the integrity and quality of markets.  From 
20   our perspective, one of the largest of these conflicts 
21   has to do with how market data is collected and 
22   disseminated.   
23             As a consequence of a broad range of quotation 
24   and transaction latencies that exist in the U.S. equity 
25   market structure, we, as fiduciary market participants, 
0114
 1   are faced with a large burden of trying to protect our 
 2   clients from the potential adverse consequences of being 
 3   in the slow lane of the market data highway. 
 4             Of all the risks that fiduciaries like us worry 
 5   about, I think it is fair to say that one of our chief 
 6   concerns about client protection should not be the result 
 7   of a risk that is within our collective power to 
 8   mitigate.  Consequently, we are supportive of more 
 9   comprehensive SRO governance reform over time.   
10             I will now turn to the subcommittee's specific 
11   recommendations for improvements within the existing 
12   structure. 
13             Regarding their first recommendation on SRO 
14   immunity from liability, we support their conclusions 
15   that liability limitations should be broadly reviewed 
16   with a goal of higher and more consistent levels of 
17   liability along with adequacy of reserves held against 
18   them. 
19             Our support for meaningful SRO liability reform 
20   is underscored by the practical reality that, due to the 
21   protected quotation status of exchanges, as was 
22   referenced earlier, market participants are routinely 
23   compelled to accept unilaterally adopted liability 
24   limitations they may not view as appropriate. 
25             We, likewise, support the subcommittee's second 
0115
 1   set of recommendations on strengthening the role of the 
 2   NMS plan advisory committees.   
 3             The frustration of the advisory committees 
 4   concerning the current limitations on their ability to 
 5   participate in and influence deliberations and decisions 
 6   of the operating committees has been well-described by a 
 7   number of commenters.   
 8             These limitations help give rise to conflicts 
 9   such as the market data issues that I mentioned earlier. 
10    They also make it harder to find well considered and 
11   designed solutions to market structure problems such as 
12   the SIP failures that took place in 2013 and '14 and 
13   market function failures like those that occurred in 
14   2010, 2012, and 2015, increasing the risk of future 
15   market failures and decreasing investor confidence in 
16   markets. 
17             The subcommittee's third recommendation of 
18   tying the timing of rule effectiveness to the release of 
19   technical information required to implement the rule 
20   makes good sense, especially given the ever-increasing 
21   technological complexity and intensity of markets. 
22             Market participants need time to implement and 
23   test system changes to be confident that their systems 
24   are functioning as intended, and we have seen the 
25   disastrous consequences of system changes that are rushed 
0116
 1   or otherwise not properly tested. 
 2             The committee's fourth recommendation, which is 
 3   aimed at reducing the duplication of regulatory efforts 
 4   across SROs is, likewise, laudable.  It seems clear that 
 5   a more simplified and centralized approach to common 
 6   regulation functions across market centers would be more 
 7   transparent, understandable, and efficient for market 
 8   participants. 
 9             Moreover, centralizing certain of these common 
10   functions will foster competition by lowering costs for 
11   current users and lowering barriers to entry for 
12   potential new competitors for providing execution 
13   services.  The current more duplicative and opaque system 
14   carries a higher and more hidden cost for the individuals 
15   and institutions that represent the end investor.   
16             In summary, we're grateful to the subcommittee 
17   for tackling these difficult issues and making 
18   substantive recommendations to improve the current system 
19   of trading venue regulation.  We also acknowledge a 
20   significant step the Commission took when it created and 
21   empowered the EMSAC to advise it on a range of 
22   challenging issues and encourage the Commission to 
23   continue this important work on reforming the current SRO 
24   and NMS plan constructs.  
25             Thank you. 
0117
 1             MR. LUPARELLO:  Thank you, David.  Jeff? 
 2             MR. DAVIS:  Yeah.  As you said, Steve, I am a 
 3   repeat participant, so I will largely rest on the 
 4   comments made this morning, except to again reiterate our 
 5   thanks and appreciation to the subcommittee members for 
 6   their hard work and their good work on these topics, to 
 7   say that we are open to a full debate about this, and to 
 8   say that we will treat them respectfully and thoroughly 
 9   in our forthcoming comment letter that -- unfortunately, 
10   we are not fully prepared to discuss them today.  Thank 
11   you. 
12             MR. LUPARELLO:  Thank you, Jeff.  
13             Brett? 
14             MR. REDFEARN:  So, do I get to use some of his 
15   time? 
16             MR. LUPARELLO:  You were going to anyway. 
17             MR. REDFEARN:  I already used it.  No, I timed 
18   it. 
19             So, thank you very much, Chair White, Steve, 
20   SEC staff, members of the committee, really appreciate 
21   you having me here again today to talk about these 
22   issues.  I'm Brett Redfearn, and I run Equity Market 
23   Structure for J.P. Morgan globally. 
24             As you know, J.P. Morgan -- I could say a 
25   number of different things, but I'll just say that we're 
0118
 1   currently representing about 80 percent of the U.S. 
 2   equity market, so we're a very large participant there.   
 3             My current responsibilities as -- in addition 
 4   to other things, is to be a representative to the 
 5   advisory committee of two of the different plans, the 
 6   limit up-limit down plan as well as the CTA/UTP plan.  
 7   So, I have ample experience in this advisory committee 
 8   function. 
 9             And I just wanted to start by just saying 
10   thanks for your hard work.  You guys have had a lot of 
11   meetings; you've gotten a lot accomplished.  The 
12   recommendations that you brought here today are great.  
13   In Market Structure we talk about things for a long, long 
14   time, so seeing tangible things on the table -- I really 
15   think this is -- it's a fantastic and tangible step 
16   towards taking this beyond the realm of debate into the 
17   realm of action, so thanks for all that good work. 
18             Specifically, Rick, your -- the subcommittee 
19   that you've been chairing -- I would just say this:  with 
20   all of the recommendations that you have there, if we 
21   approved every single one of them as is, we would take a 
22   meaningful step forward.  So, just as a starting point, 
23   we think that that -- that they're all good and they're 
24   all meaningful, and they all are headed in the right 
25   direction.  So, you know, if we got that done, we would 
0119
 1   be farther than we certainly have been on this for a 
 2   while. 
 3             Not surprisingly, as you might --  
 4             A PARTICIPANT:  There's a but there, so --  
 5             MR. REDFEARN:  So, you said to me earlier -- I 
 6   know what you're probably going to say, and you're going 
 7   to say, "But you don't go far enough."   
 8             So, yes, you know, there are some areas where 
 9   we think that there might be some tweaks that could be 
10   done and one particular area where we think it doesn't go 
11   far enough.  So, even though I made some comments to this 
12   the last time that I was here, I just want to reiterate 
13   that in a second. 
14             But first, I wanted to make a couple comments 
15   about SRO immunity and limited liability.  First of all, 
16   we appreciate the language of rule-based liability 
17   limits.  We think that this does convey that liability 
18   should be narrowly applied to rule-based attributes of 
19   exchanges and not to all of the diversified aspect of 
20   exchanges, so good choice of words there. 
21             Secondly, rule-based liability levels should be 
22   increased.  I think that's -- I think everybody kind of 
23   recognizes that.  These are very old limits.  They don't 
24   really represent this scope of activities and the amount 
25   of business in our market today and the world that we're 
0120
 1   in.   
 2             Substantial increases, the -- you know, I think 
 3   probably the Facebook IPO experience is a very good 
 4   example where hundreds of millions of dollars turned out 
 5   to be losses to the broker-dealer community.  So, this 
 6   just gives you some sense that we're a long way from 
 7   where we should be. 
 8             They should be meaningfully increased.  I'm not 
 9   sure what the numbers are.  I'm sure in the subcommittee 
10   you've debated this.  It's got to be at least 10, 20 
11   times where they are today, if not more.  So, maybe 
12   formulaic, so you do have this differentiation between 
13   the activity of primary versus non-primary markets.   
14             And also, regulatory capital is a good idea.  
15   If there is a big event, having regulatory capital on 
16   hand reduces some of that risk.  So, we think that that's 
17   an important thing, and we should move forward with that. 
18             By the way, broker-dealers are subject to these 
19   type of capital requirements, ECNs, before becoming 
20   exchanges, have familiar with the level of regulatory 
21   capital that is required. 
22             Secondly, I see the subcommittee did not reach 
23   a consensus on the issue of regulatory immunity.  We just 
24   think that this is something you should stick with, you 
25   know.  Don't give up on this one too easily.  There 
0121
 1   really should be a distinction between the regulatory 
 2   functions that warrant regulatory immunity and the other 
 3   exchange businesses that are not.   
 4             We all have seen diversification, new business 
 5   lines, new acquisitions, things that go way beyond what 
 6   we think of as core exchange services.  So, we need to 
 7   make some differentiation between those and apply this 
 8   immunity concept appropriately.   
 9             So, we would certainly argue that you keep at 
10   that one and sort of figure out where that line is.  I 
11   know it's not an easy one, but this is work that's 
12   important because, ultimately, as they -- as exchanges 
13   compete with brokers and other members of our economy in 
14   different areas, having immunity really is an unfair 
15   competitive advantage that should be addressed. 
16             Given the time constraints, I'm not going to 
17   make any specific comments on recommendations three and 
18   four at this time except to state we're fully supportive. 
19             So, let me move on to my favorite, which is NMS 
20   plan governance.  So, first of all, I think you read 
21   these and it reflects a few things, right.   
22             It reflects the fact that, one, there are 
23   problems with the selection process of advisory committee 
24   members.  Two, advisory committee impact is inadequate in 
25   the current plan decision-making.  Three, there is 
0122
 1   excessive and inappropriate use of executive sessions 
 2   that further disenfranchises advisory committee members. 
 3    And four, the voting construct does not work as it 
 4   should, right.  So, those things all come through when 
 5   you read these recommendations. 
 6             Before addressing the specific ones, I just 
 7   wanted to start with the one thing that's not included in 
 8   the recommendations, which is direct voting 
 9   representation.   
10             You know, we talked about this the last time.  
11   Looking at this, I understand that it's hard to get 
12   there; it's hard to get to this point of having direct 
13   voting representation.  But I don't -- I'm not sure why 
14   we can't get there.  I think, if we don't include that, 
15   we're at risk of missing a real opportunity here.  I 
16   don't know when we're going to come back around and do 
17   this again.  And so, having this opportunity now, I think 
18   we need to give this extremely hard look. 
19             And when you think about voting 
20   recommendations, I don't understand why, in light of the 
21   following, right -- we know that exchange boards have 
22   voting representation and exchange boards function very 
23   well.  We know that the FINRA board has broker-dealers 
24   with voting representation, and I think Rick would attest 
25   that the FINRA board functions pretty well.  The DTCC 
0123
 1   board has broker-dealers as representatives to the board, 
 2   and that functions very well. 
 3             NMS plans do not have voting representation of 
 4   broker-dealers or asset manager representations.  And I 
 5   don't think they do function particularly well.  The 
 6   results are often less representative, less effective, 
 7   and more contentious than is necessary. 
 8             Now, on this, I know that Joe and Bats had 
 9   supported a vote for brokers and asset managers on the 
10   SIP plans and they had taken something there.  So, we 
11   appreciate that level of support.  We thought it was a 
12   good one.  It really didn't get much of a hearing in the 
13   -- in that committee. 
14             Now, just to illustrate a few of the points 
15   about why this is so important, the SIP plans today are 
16   solely governed by SROs that have clear and obvious 
17   conflicts of interest at the table, right.  Again, last 
18   time I talked about this.  I'll just reiterate a couple 
19   of the specific ones that I mentioned. 
20             But the first one is that the voting members of 
21   the SIP plans are selling market data products that 
22   compete with the product of the SIP.  They're selling 
23   competing products, so that's a pretty clear conflict. 
24             The second one is, the voting exchanges of the 
25   SIP plans have a disincentive, a natural disincentive to 
0124
 1   invest in the SIP plans to make them competitive 
 2   products.  As we know, the more money that's spent on the 
 3   SIPs, the less revenue that's left over for divvying up 
 4   among the exchanges after you take out cost, right.  So, 
 5   the more you spend, the less that's left over to be 
 6   shared. 
 7             Also, the better that the SIPs function, the 
 8   less competitive the proprietary products are relative to 
 9   the SIP.  So, there are certain things where we could 
10   actually use a SIP for in terms of trading products.  But 
11   having it in the shape it's in today means that we can't 
12   even use it for that.  So, you have to then buy 
13   proprietary products, and we all know that those are huge 
14   revenue drivers. 
15             So, exchanges are not only economically 
16   incentivized to -- you know, for the SIPs competitive, 
17   they're the only ones with such clear conflicts.  And so, 
18   it's surprising that they're the only ones that have a 
19   vote, which I will sort of come back and pose a question: 
20    in light of this, why can't we consider truly broadening 
21   out the vote. 
22             The suggestion has been made that brokers 
23   wouldn't have the same obligations to NMS plans that 
24   exchanges have because we're brokers and we're not 
25   regulated by the SEC.  But I would argue this:  When we 
0125
 1   go to the SIP meetings, we are there because we want to 
 2   make them competitive; we want to make them work; we want 
 3   to make them usable.  
 4             In light of all of that, our recommendations 
 5   have been about adding transparency, adding redundancy, 
 6   improving speed, improving quality.  So, again, I say, 
 7   you know, why would it be that we don't have the same 
 8   obligations as the plan.  I think our intentions and our 
 9   motives are good ones, and these are the incentives that 
10   you should want among voting participants to the plan. 
11             I would also argue that, when we talk about 
12   obligations as a voting member of the plan, if the 
13   existing participants have not upheld their 
14   responsibility to the plan -- and what I'm saying is that 
15   -- their task to provide prompt access to market data. 
16             For most broker trading systems, the SIPs are 
17   of limited use, and they primarily serve as backups, 
18   right.  They primarily serve as backups to the superior 
19   proprietary data products. 
20             In the most recent data that was provided, the 
21   average quote latencies of the SIPs are 900 microseconds 
22   solely to aggregate a quote.  That doesn't include all of 
23   the unnecessary geographic travel time that every quote 
24   has to take around the state of New Jersey before it 
25   comes to wherever it needs to go.  
0126
 1             And so, there's a significant amount of 
 2   latency.  And if anyone doesn't think that 900 
 3   microseconds of time is a long period of time, I would 
 4   recommend that you read some of the IEX comment letters, 
 5   which suggest that 900 microseconds is a real long time 
 6   and quotes shouldn't be provided order protection if it's 
 7   taking that long. 
 8             So, there's a very uncomfortable inconsistency 
 9   here when we talk about what is timely, what is prompt, 
10   and what is this debate that's going on. 
11             The economic incentives for exchanges in 
12   today's data landscape really revolve around proprietary 
13   data products.  This is where you've seen innovation.  
14   This is where you've seen significant investment being 
15   made, and this is where we have products that really 
16   serve trading systems in the markets that we have today. 
17             They also happen to be characterized by 
18   unconstrained fees and unconstrained -- and increasing 
19   fees, right.  So, we've all seen sort of this escalation. 
20    There was some reference that was made to that earlier 
21   today, very in contrast to the SIPs and the consolidated 
22   data feeds, which are -- which have antiquated 
23   architecture and inadequate consolidation process and 
24   some questionable process practices regarding the 
25   purchase and utilization of bandwidth that may result in 
0127
 1   even further delay.   
 2             So, just basically, the SIPs are delayed data 
 3   being sold as real-time data at a cost of nearly $400 
 4   million a year to the industry in the current setup.  And 
 5   as -- I would say we either need to fix the SIPs or phase 
 6   them out.  And I know there's another whole discussion 
 7   you guys are going to have on the whole market data 
 8   question.  However, I would simply say that, if we don't 
 9   fix governance model, it's unlikely that we're truly 
10   going to fix the SIPs the way that we want to today.  So, 
11   the incentives are simply not there. 
12             I would just simply close by -- close on this 
13   point by quoting Tom Farley and what was said earlier 
14   about another committee, which was we urge the SEC not to 
15   give up but to start anew with a more representative 
16   committee, talking about a different committee, however, 
17   in this case. 
18             So, with that being said -- you guys encouraged 
19   me to go ahead with that one.   
20             So, before concluding, let me just make a few 
21   comments, Rick, on the specific points of the committee, 
22   right.  First of all, with respect to clarifying the 
23   process for selecting advisory committee representatives, 
24   I don't think they should be selected solely from the SRO 
25   participants.  We've seen in the past that, even if non-
0128
 1   SROs recommend names, if somebody is brought before the 
 2   committee and they're potentially going to be a 
 3   controversial individual or something like that, they can 
 4   be voted down.  We've seen that happen, so very cautious 
 5   about solely having this being brought forth by the SROs. 
 6             I think that representation by -- coming from  
 7   -- you know, whether it's SIFMA or STA or ICI or some of 
 8   the different industry groups -- might be a good way to 
 9   try to ensure that there's direct industry representative 
10   coming from these involved trade associations. 
11             Secondly, in terms of expanding and formalizing 
12   the role of the advisory committee, again, great proposal 
13   in conjunction with the vote. 
14             Thirdly, with respect to the recommendation to 
15   significantly narrow the use of executive sessions, this 
16   is absolutely necessary, and it would help.  It would be 
17   a huge help, great recommendations.  We've brought this 
18   up many times in the CTA/UTP meeting, and so far it's 
19   still very kind of subjective and nebulous in terms of 
20   what are the causes for executive sessions and what 
21   actually happens in there. 
22             Fourthly, with respect to the working groups, 
23   I'm going to differ on one point.  I think that it's 
24   missed the target in terms of advisory committee members 
25   not having representation on the working groups.  Our 
0129
 1   concern here is that, if you tighten up the use of 
 2   executive sessions and then you have working groups where 
 3   advisory committee members aren't part of it, we're 
 4   concerned that the working groups actually become sort of 
 5   the de facto place where all the meaningful content 
 6   discussions happen and the operating committee comes back 
 7   to a place where it's more of a rubber-stamping kind of 
 8   body where you're not getting into the meat of it. 
 9             The limit up-limit down plan has actually 
10   recently expanded into allowing advisors to be on the 
11   working groups, and it's been extremely productive.  So, 
12   I think that's an example of showing how that could 
13   indeed work. 
14             With respect to the unanimous vote, limiting 
15   that, I think that's a great idea.  I'm not sure if you 
16   meant to say, "limit the unanimous vote," or, "eliminate 
17   the unanimous vote".  Was it intended to sort of 
18   eliminate unanimous voting? 
19             MR. KETCHUM:  Effectively eliminate the --  
20             MR. REDFEARN:  Yeah.  So, I think we just argue 
21   eliminating.  No one participant really should have veto 
22   power over the rest of the committee. 
23             And last but not least, with respect to 
24   revisiting the allocation of voting rights among SROs, 
25   we're supportive.  We think that that generally makes 
0130
 1   sense.  We would only ask that you also consider the 
 2   issue of if there's an equities exchange that is no 
 3   longer -- if there's an exchange that was sort of still 
 4   on the committee that no longer runs an equities exchange 
 5   nor ever plans to run an equity exchange again.   
 6             We really need to figure out how to deal with 
 7   that voting issue that's been brought up.  And it hasn't 
 8   been resolved yet, it's sort of lingering.  So, after 
 9   some point in time, if somebody's not planning on being 
10   an equity exchange, they probably shouldn't have a vote. 
11             That's all I got.  Thank you very much for your 
12   time.  I look forward to discussing with you more. 
13             MR. LUPARELLO:  Thank you, Brett.   
14             Perhaps people need a second or two to let 
15   Brett catch his breath and --  
16             MR. REDFEARN:  I know I do. 
17             MR. LUPARELLO:  And I will open it up to the 
18   committee.   
19             MR. RATTERMAN:  Just to make one point in 
20   response to Brett's thoughts on the limited liability and 
21   how the exchange model should be looked at again, we're 
22   certainly supportive of a visit of that.   
23             But I think one thing that needs to be made -- 
24   one point that needs to be made clear is that, unlike 
25   broker-dealers that can negotiate with clients on a one-
0131
 1   on-one basis and potentially turn customers away all 
 2   together and respond to customers of different sizes, 
 3   exchanges do have a fair access obligation to all 
 4   elements of how exchanges are run.   
 5             And so, the exchange liability limits need to 
 6   be looked at in light of that very real limitation that 
 7   anybody who shows up at the exchange's door is going to 
 8   be allowed access and under the same commercial 
 9   relationship that any other customer is.  And that is a 
10   unique model difference between the exchanges and fair 
11   access versus broker-dealers.  
12             So, we're certainly supportive of looking at 
13   the liability limits and trying to find more consistency, 
14   but some of the sizes of those limitations that have been 
15   proposed, I think, may not take into account the fair 
16   access real limitation that exchanges have.   
17             MR. REDFEARN:  Do you think that applies, Joe, 
18   to all of the different functions of the exchange, or do 
19   you think there -- I mean, because with the exchanges 
20   diversifying, do you think at some point there are 
21   certain things where those liability limits are more 
22   applicable than other areas? 
23             MR. RATTERMAN:  The main point I'm making is 
24   that the liability limits than an exchange would have 
25   with each and every participant have to be applied on a 
0132
 1   fair access basis.  And that's the difference between 
 2   broker-dealers that can negotiate different terms with 
 3   different clients.  And so, the liability limits have to 
 4   be thought of in light of the fact that those limitations 
 5   have to be applied fairly and universally to every single 
 6   client that shows up.   
 7             So, it's more of a commentary on the 
 8   relationship between the exchange and its participants, 
 9   not on the rule-based or activity that's in question.   
10             MR. LUPARELLO:  Manisha and then Maureen. 
11             MS. KIMMEL:  Question for Brett on the voting. 
12             Assuming that the advisory committee had one 
13   vote, how much would it matter given that it would be a 
14   minority -- you know, it would be the only one? 
15             MR. REDFEARN:  So, I would ask -- you know, ask 
16   Rick or ask the Chicago stock exchange or any other 
17   participant at the table who only has one vote how they 
18   feel about one vote.   
19             I mean, we don't think that we're going to come 
20   in with a majority.  We just think that you have a 
21   negotiation and a stature and some influence that goes 
22   beyond, you know, sort of -- I mean, as it is now, you 
23   throw an input, and then, you now, you're kind of shut 
24   out of the room when votes take place and the final 
25   decision is made.  Just -- a vote is just a much more 
0133
 1   meaningful part of the governance process. 
 2             MR. KETCHUM:  Maybe to follow up on Manisha's 
 3   question to that -- I mean, far be it for me to come out 
 4   against votes in the District of Columbia.  But I -- 
 5   honest answer, I think FINRA values the fact it has a 
 6   free and open opportunity to interact from the standpoint 
 7   of the plan.  No, obviously, we wouldn't give away our 
 8   vote if we had a choice, but we do get outvoted in a 
 9   variety of circumstances.  So, I don't think we think the 
10   vote is terribly relevant. 
11             What we do think is quite important is the 
12   ability to be very much involved and, as you say, have 
13   the respect of being very much involved in formulating 
14   responses to problems.  
15             So, taking your points in the working group, 
16   which I think the subcommittee clearly will have to 
17   reflect on, it doesn't seem like it's the vote that's 
18   going to make a difference.   
19             It seems like your concern is being shut out of 
20   an effort to try to both issue, spot, identify things 
21   that you feel the industry may have different views than 
22   the SROs on -- from the standpoint of importance in one 
23   way or another and that that would seem to be most of the 
24   areas where your ability to make an impact is harder, at 
25   least by the construct that you set up. 
0134
 1             But am I missing something, Brett, from that 
 2   standpoint? 
 3             MR. REDFEARN:  I mean, I think I heard you said 
 4   that you -- FINRA wouldn't want to give up their vote.  I 
 5   mean, there's something about having a vote --  
 6             MR. KETCHUM:  I also said we don't think it's 
 7   very important. 
 8             MR. REDFEARN:  You know, so, I mean, look, 
 9   there may be issues from time to time where -- by the way 
10   we also think that asset managers should be -- have 
11   voting representation as well.  And there may be a point 
12   in time where an asset manager, a broker, some of the 
13   exchanges -- that, you know, you actually have a 
14   demographic in a committee where it's meaningful in terms 
15   of a decision and that, by being part of that, then 
16   there's a negotiating power that we would have that we 
17   simply don't have.   
18             And that creates a level of input and influence 
19   that just doesn't exist; whereas, today it's sort of like 
20   you can say whatever you want, but if you don't have a 
21   vote, then it's kind of irrelevant when push to shove 
22   because you're not the one who they're negotiating with 
23   to try to get something passed.   
24             So, I mean, I would ask you why, if it's -- you 
25   know, if it's not that big of a difference, then why 
0135
 1   would we not be able to proceed with offering a vote? 
 2             MR. KETCHUM:  Well, I -- probably a 
 3   clarification's important.  The subcommittee didn't reach 
 4   a determination that there shouldn't be a vote.  It 
 5   didn't reach a determination either way.  It instead 
 6   focused on a variety of recommendations that it thought 
 7   would make the advisory committee and the participants of 
 8   the advisory committee truly be participants in the 
 9   process. 
10             So, it just -- it shouldn't be read -- and it 
11   would be unfair because I think there were different 
12   views among the committee -- that a determination was 
13   made that there shouldn't be votes on the rest.   
14             I think there was a determination made that it 
15   was more -- somewhat similar to what Joe described this 
16   morning, it was more complex, given the nature of the 
17   exchange rule and given the nature of the plan rule that 
18   specifically anticipated that all members would be -- of 
19   the executive and operating committee would be -- and 
20   members of the plan would be SROs.   
21             So, the effort was to try to identify a 
22   structure where there would be effective participation 
23   within the construct of the plan rule.  But there wasn't 
24   a decision one way or another as to whether a vote and 
25   specific participation might be appropriate.  It just -- 
0136
 1   the subcommittee didn't reach that point. 
 2             MR. REDFEARN:  The plan rule being the way that 
 3   the plan is currently approved, or are you talking about 
 4   the Exchange Act part of it? 
 5             MR. KETCHUM:  The SEC's plan rule. 
 6             MR. REDFEARN:  Yeah. 
 7             MR. KETCHUM:  And interesting questions from 
 8   the standpoint of the statute. 
 9             MR. LUPARELLO:  Maureen? 
10             MS. O'HARA:  I just wanted to follow up on the 
11   comments you made about the SIP.  We really have not, as 
12   a group really discussed that.  And I think in part it's 
13   because you came up with a long list of problems with the 
14   SIP, but actually I think there's far more and that, if 
15   we all began to add in our concerns with the SIP, it 
16   would be a long time.   
17             I mean, as an academic, one of the challenges 
18   we have trying to do research is, the quality of the SIP 
19   with respect to time stamps, for example, has been a huge 
20   problem.  And again, we can thank the SEC for beginning 
21   to move forward on those sorts of issues.   
22             So, I can assure you that I think the SIP 
23   issues will be ones that we will be devoting a lot of 
24   time to down the road, and your comments are very 
25   helpful. 
0137
 1             MR. LUPARELLO:  Gary? 
 2             MR. STONE:  So, Brett, the way I looked at it  
 3   -- and this is my view -- when we were talking about it 
 4   at the subcommittee was, the way we try to -- the way I 
 5   thought we were -- construction was the advisory 
 6   committee now has a full role whereas it didn't before.  
 7   So, they're a bloc within themselves. 
 8             And what we thought of is, what happens when 
 9   the advisory committee, which is maybe going to be a 
10   minority on it, gets outvoted by the operating committee 
11   or the members in a straight-up vote. 
12             Well, what we thought was, really the 
13   commission should know that the advisory committee 
14   doesn't agree.  And this was a way or a process of coming 
15   to the Commission and saying, "The operating committee 
16   wants to do this," making sure that the Commission 
17   understands what it is documented; the advisory committee 
18   disagreed with it; and have the Commission understand or 
19   have the transparency to understand that there's a 
20   difference of opinion.  And that way they can make the 
21   decision and perhaps come to a different conclusion. 
22             So, we thought of it as actually elevating even 
23   higher the role of the advisory committee in this 
24   construct. 
25             MR. REDFEARN:  I mean, as I said before, I 
0138
 1   think it's a positive step forward.   
 2             I don't know if I was supposed to respond to 
 3   that. 
 4             MR. LUPARELLO:  No, no, absolutely. 
 5             MR. REDFEARN:  Yeah, it's definitely an 
 6   improvement.  It's a step forward.  It's a good thing.  
 7   We -- like I said, we just don't think it goes far 
 8   enough.  And I'm sort of throwing the question back of 
 9   what is the compelling reason why brokers shouldn't 
10   actually have a vote.  You know, we're not going to carry 
11   a majority. 
12             MR. STONE:  But that's why, because you're not 
13   going to have -- brokers won't be able to carry the 
14   majority.  So, how is it that you actually elevate the 
15   minority to a position where the ultimate arbiter is 
16   going to be the Commission to understand there's a 
17   difference of opinion and --  
18             MR. KETCHUM:  This might be a good time to, 
19   since Jeff's had the mike on, to let him speak on it, 
20   too. 
21             MR. DAVIS:  So, I should start by saying that 
22   we are -- NASDAQ is open to all these recommendations 
23   about plan governance.  I would hope that Brett, as an 
24   active member of the advisory committee now, would 
25   acknowledge that, even before this committee was 
0139
 1   convened, we have begun making process in trying to 
 2   respond to the industry, to Brett. 
 3             We have more advisors on the NMS committees 
 4   that I have attended, the CTA and UTP committees, than 
 5   ever in the past.  We have created specific agenda items 
 6   to hear from the advisors.  We have created specific 
 7   minutes and other communications to make sure that we 
 8   respond to their concerns and that we address their 
 9   concerns and state the reasons why we agree or disagree 
10   or respond to factual questions.   
11                  We have made a concerted effort, again, 
12   even before this committee, before any commission action, 
13   to reduce the number of items that are discussed in 
14   executive sessions.  We have explained to the advisors 
15   that the reasons we still have executive sessions are 
16   when we are talking about litigation, when we are talking 
17   about specific data vendors or specific broker-dealer 
18   where you might not want to be discussing that with 
19   competitors.   
20             And we have said that we're open to further 
21   improvements to make sure that the industry feels like it 
22   is well represented and well served by the operating 
23   committee.  So, I mean, I hope that the advisors 
24   acknowledge that we have made progress in that area and 
25   that we have committee to continue making progress.   
0140
 1             I think the difficulty for us with the vote 
 2   partly revolves around the -- you know, how much of a 
 3   vote is really necessary for it to become relevant, and 
 4   that number really worries us because we have serious 
 5   obligations under the Exchange Act.  We have serious 
 6   obligations under the SEC rules that apply only to 
 7   exchanges.   
 8             We understand that broker-dealers are subject 
 9   to SEC regulation and that they do have, you know, 
10   responsibilities, but they are different from ours.  And 
11   they are not always consistent with ours, and we are 
12   often regulating their conduct under these plans.   
13             And you know, I don't know if the subcommittee 
14   intended to distinguish among the plans.  That would be 
15   one of the complexities we think has to be thought 
16   through.   
17             But there are many different NMS plans.  The 
18   Commission looks to the exchanges to accomplish a lot of 
19   what it sees as important to the marketplace.  It is more 
20   than just data distribution.  It is more than just data 
21   revenue collection as some people like to portray.  It is 
22   limit up-limit down; it is the tick pilot; it is going to 
23   be CAT.  There are many, many different kinds of NMS 
24   plans.  They have many, many different kinds of 
25   functions.  And we think that you have to have some 
0141
 1   nuance when you talk about what the roles are and what 
 2   the obligations are and what the conflicts are. 
 3             And so, I would also point out that to the 
 4   extent that we are conflicted in the way that we operate 
 5   these plans, that is just endemic to our industry.  All 
 6   of our regulatory efforts in many, many different ways 
 7   are about controlling conflicts.  We think that there are 
 8   mechanism in place that control the conflicts that are in 
 9   place here.  And I think the evidence of that is very 
10   clear.   
11             Even -- again, before we get to what this 
12   committee does, we have announced months ago that we are 
13   launching a new SIP.  We invited the advisors in to hear 
14   about the proposals, to discuss the proposals, to 
15   understand what we were accomplishing.  And come the 
16   fourth quarter, the new SIP for the NASDAQ UTP will have 
17   latency as low as 50 microseconds.  I guess I would have 
18   to correct the record.  It is not 900 microseconds on 
19   average; it is between 4 and 450 microseconds.  And that 
20   is all up on the website. 
21             But nonetheless, I mean, we are working to 
22   improve this.  We are working to address the industry's 
23   concerns.  We, you know, have heard and are trying to 
24   respond, and we feel like that is already built into the 
25   process. 
0142
 1             MR. KETCHUM:  Maybe one question I could ask, 
 2   Brett as well -- or the committee, the panel itself, what 
 3   are the difficulties of exchange participation or 
 4   industry participation in the plan as a question of 
 5   accountability?   
 6             There are sort of two parts to plans.  One is 
 7   the ability to fashion the plan and present it to the 
 8   Commission for approval.  The other is that the 
 9   Commission's plan rule and the -- or the Commission's 
10   actions around it impose an accountability to produce 
11   something, CAT being perhaps the most current example of 
12   something where the Commission can hold the exchanges and 
13   FINRA accountable for failing to produce at some point a 
14   CAT plan. 
15             How would that -- since, obviously, the entire 
16   industry can't be part of the plan, how would that work 
17   from the accountability side, Brett, from the standpoint 
18   of trying to push towards the range of compromises 
19   necessary to deliver a proposal to the Commission? 
20             MR. REDFEARN:  I mean, you have -- when you 
21   have a more reflective set of voices with votes 
22   representing the industry as a whole, there is almost a 
23   checks and balances sort of thing that's built in there. 
24             You know, the point I made earlier was that the 
25   SIPs today, when people talk about all the issues, if 
0143
 1   there was an obligation to the plan or an obligation or 
 2   responsibility or accountability, didn't happen.  It is 
 3   clearly reflecting that it didn't work.  It hasn't 
 4   worked.  And so, when we look at how slow it is relative 
 5   to the private market, it hasn't been accountable.  If 
 6   that was the case, we wouldn't be having this 
 7   conversation. 
 8             So, what we have now isn't working.  And when 
 9   you think about the SIP and the situation like this, like 
10   I said, I mean, we're there.  And I mean, we brought up 
11   an issue to the committee last time, why aren't we 
12   considering having all of the feeds going to Mahwah 
13   Carter in Secaucus directly so you don't have all of this 
14   geographic latency in the mix?   
15             The private market did that 10 years ago.  And 
16   we throw that on the table, and you know, you kind of get 
17   a bit of a cross-eyed look, and it's discussed in 
18   executive session.  Maybe we'll talk about it again.  
19             But this is something that a long time ago I 
20   think that we probably should have been talking about, 
21   and it's not happening.  I think it's because it would be 
22   expensive; it would potentially compete with a prop data 
23   model, and it just hasn't been on the table.  And we have 
24   to do something to create -- I think it makes it more 
25   accountable actually. 
0144
 1             Does that make sense, Rick? 
 2             And by the way, it is 920 microseconds on 
 3   average.  The median is around 400.  But the Q1 2016 
 4   numbers for the UTP SIP just came out.  It was -- I think 
 5   920 mics on the website. 
 6             MR. LUPARELLO:  Now that we've discussed the 
 7   difference between mean and median --  
 8             MR. REDFEARN:  Average.  I meant average. 
 9             MR. LUPARELLO:  Rick, I guess I would suggest 
10   maybe, having listened to this conversation -- first of 
11   all, I should make sure.   
12             Is there anyone else on the committee who has 
13   questions or comments or observations? 
14             Hearing none, maybe I would suggest that you 
15   follow the path that Kevin and his subcommittee have 
16   followed.  I also think -- and this is speaking just for 
17   me.  Obviously, the subcommittee should do whatever it 
18   thinks best -- that these are four very significant 
19   substantive recommendations.  They are of a general 
20   category, but they are -- they do have different specific 
21   elements.   
22             If there is one or two in particular you want 
23   to push forward as more concrete recommendations, you 
24   could certainly consider that.  If you want to continue 
25   forward on all four, obviously, that's the subcommittee's 
0145
 1   choice. 
 2             I think, listening to this conversation, also 
 3   listening to other conversations and just the views of 
 4   the staff generally, I think the second recommendation is 
 5   the one that is the most immediate -- and perhaps the 
 6   immediate of them, but just that's something for the 
 7   subcommittee to take back and consider. 
 8             So, if that's -- if we can work out a time 
 9   schedule where there is a formal recommendation from the 
10   subcommittee that can be made public in the next couple 
11   of weeks with a public committee meeting a couple of 
12   weeks after that, is that something the subcommittee can 
13   meet? 
14             MR. KETCHUM:  We'll certainly try to have that 
15   conversation and move forward as quickly as we can, yes. 
16             MR. LUPARELLO:  Very good. 
17             Well, we actually gained a little time.  I'll 
18   thank Jeff again for that. 
19             And why don't we take a 15-minute break and 
20   reconvene at 2:00 for the other subcommittee updates.  
21   Thank you.   
22             (A brief recess was taken.)  
23             MR. LUPARELLO:  Last order of business for 
24   today is an update from the Market Quality Subcommittee 
25   and the Customer Issues Subcommittee.  I'm going to ask 
0146
 1   each chair of the subcommittee to provide a brief report 
 2   on the issues and potential recommendations their 
 3   subcommittee are developing.  And I welcome comments and 
 4   questions from the committee.   
 5             I would also acknowledge that these committees, 
 6   as the Chair pointed out, have worked extremely hard and, 
 7   in fact, are very well along on some quality, substantive 
 8   recommendations.  And we, unfortunately, lacked the time 
 9   and capacity to deal with them and give them the 
10   attention they deserve today but would, again, recognize 
11   just how much work those committees -- those 
12   subcommittees have done.  
13             So, with that, can we start with the Market 
14   Quality Subcommittee?   
15             And Eric, I'll ask you to give a brief readout.  
16             MR. NOLL:  Thank you, Steve.   
17             As Steve said, we have had a number of 
18   meetings.  We've been working pretty hard on some of 
19   these issues, many of them keyed off of August 24th, but 
20   we are taking -- or trying to take a much broader 
21   perspective as well, around some of those issues.  
22             Like the other subcommittees, we have tried to 
23   be inclusive of market participants, so we have had 
24   representatives of the options industry.  Both the CBOE 
25   and the ISE joined our meetings.  We've had exchange-
0147
 1   traded ETP issuers in our meetings.  Bats has been 
 2   attending our meetings as well, as well as other 
 3   organizations like Healthy Markets; Weild & Company; 
 4   several market-making firms; S&P/Dow Jones as an index 
 5   provider; and ETF market makers.   So, we've tried to 
 6   have as wide of an inclusiveness as possible of people on 
 7   our committee talking about some of the issues.  
 8             The recommendations where we're ending up are 
 9   essentially around four areas.  We will continue to 
10   refine those, and we'll continue to work on those and 
11   have them more substantively presented at the next formal 
12   meeting. 
13             The first area really is about limit up-limit 
14   down.  So, when we've looked at limit up-limit down, 
15   we've come to the perspective that, from a macro point of 
16   view, the limit up-limit down mechanism does what it's 
17   intended to do sort of in a macro respect, which is to 
18   prevent runaway stocks, to prevent panic selling or panic 
19   buying, and actually introduces a pause into the market, 
20   which generally we think is a good thing so the events 
21   that go back to the flash crash do not happen under 
22   today's limit up-limit down scenario. 
23             The area that we've been more focused on around 
24   that is the reopening up of trading after a limit up-
25   limit down halt.  And so, we've spent a little time 
0148
 1   looking at what the auction can do or why doesn't it 
 2   work.   
 3             And we have reached the conclusion or at least 
 4   a majority of the committee at this point has reached a 
 5   conclusion that the auctions are not really fixable under 
 6   the current market structure and design around 
 7   competitive different environments.  And there's a bunch 
 8   of reasons for that about why they don't work very well 
 9   during the intraday trading day. 
10             So, rather than try to struggle to fix the 
11   auction in some way, our recommendation is to eliminate 
12   the auction and have the limit up-limit down band slide 
13   over time. 
14             So, as an example, a stock that hits a limit 
15   down condition would be in that limit down condition, 
16   say, for roughly 30 seconds.  And then, the bands would 
17   adjust around that new limit down condition, sliding up 
18   and down as they go along. 
19             The minority view, however, on the committee -- 
20   and there was one -- said that after -- either after 
21   several limit conditions or under other circumstances, we 
22   should, in fact, halt the stock and have a reopening 
23   auction.   
24             Their suggestion was that we use a volume-
25   weighted reopening process, so, as opposed to a time-
0149
 1   based one that we use a number of participants, that the 
 2   reopening would take place on a certain number of shares. 
 3   So, a stock would remain in a halted condition until that 
 4   reopening number of shares showed up and the stock could 
 5   be reopened. 
 6             The objections to that are, one, you could see 
 7   a stock never reopen in a single trading day, that those 
 8   conditions never got met to reopen the security.  It may 
 9   have -- unfortunately, have a counter effect to bringing 
10   calmness; it may introduce more panic, as a matter of 
11   fact, because if a stock couldn't open in a minimum 
12   period of time, people would then suggest something was 
13   really wrong with that company or that security and 
14   therefore could put it under additional stresses and 
15   strains. 
16             And it still does not fundamentally address 
17   some of the issues about why intraday auctions in 
18   competing markets don't work very well. 
19             The other area that we're going to recommend 
20   this is that the limit up-limit down process, regardless 
21   of where it ends up, have a concept of mean reversion 
22   built into it.  
23             So, one of the things we saw consistently 
24   around the date of August 24 and we see regularly during 
25   the trading day -- again, if a stock's trading for a 
0150
 1   hundred, trades down a limit condition at 90, oftentimes, 
 2   the very next print after that, after it reopens, wants 
 3   to be back at 100.  But it would hit a second halt on the 
 4   way up because the bands got readjusted. 
 5             And so, somehow we introduced the concept that, 
 6   if a stock is going back to where it started, that it 
 7   have that concept of mean reversion built into it so you 
 8   don't get those extra halts that are triggered. 
 9             Moving on to some other issues -- and I think 
10   this one's non-controversial.  We believe fully that the 
11   clearly erroneous rules ought to be consistent with the 
12   limit up-limit down bands.  I know that's a process 
13   that's already underway, but the subcommittee fully 
14   recommended that -- and that, if a trade takes place 
15   within the bands, it stands, and if it takes place 
16   outside the bands, it could be subject to clearly 
17   erroneous breaks at that point. 
18             An interesting topic came up around market-wide 
19   circuit breakers.  So, one of the things that we did 
20   consider at the time was whether we should move away from 
21   using a cash-based index or the cash-based S&P 500 to 
22   using the futures.  An again, looking at the data from 
23   August 24, it's because the cash index was not being 
24   calculated.  It was also delayed because of limit up-
25   limit downs and the failure of New York to be open in a 
0151
 1   timely fashion in that morning; where as the futures were 
 2   probably reflecting the real value of the S&P 500.   
 3             Generally though, the feedback that we've 
 4   gotten from the bulk of the people we spoke to and the 
 5   committee members is, we actually didn't want to halt 
 6   that morning, which raises some very interesting 
 7   questions about market-wide circuit breakers in general, 
 8   which is the general view for most of the market 
 9   participants was, thank God we didn't have a halt because 
10   that would have made things so much worse. 
11             So, that raises the interesting question as to 
12   whether the bands are correctly drawn, you know, whether 
13   they're too tight, or whether there are some unique 
14   triggers that ought to be considered before a halt is 
15   called. 
16             So, I think generally the subcommittee ended up 
17   saying we should continue to use the cash base index as 
18   the measure for market-wide circuit breakers, but we want 
19   to have some more conversations about are they designed 
20   and operating in the correct way. 
21             The last area we talked about was market 
22   openings.  And without opining on any one exchange's 
23   market model or business model about how to open stocks, 
24   we had a pretty firm recommendation, which I think will 
25   remain as we present more formally later, that all 
0152
 1   markets ought to open at 9:30 so that there -- we found 
 2   very little reason or justification for a market not to 
 3   open promptly at 9:30.   
 4             There may be unique circumstances around 
 5   individual names, having to do with news or something 
 6   else that may be an exception to that rule.  But 
 7   generally, our recommendation is markets should open at 
 8   9:30.  We believe that would have had a great deal to do 
 9   with helping mitigate all of the ETF market limit up-
10   limit down halts during the day, the other halts during 
11   the day, and the index calculation. 
12             So, some of the problems that we're trying to 
13   address with the limit up-limit down fix would have been 
14   not visible if all the market had been open in a timely 
15   fashion. 
16             The last things -- the other issues we're going 
17   to start to tackle is market-making standards, rights, 
18   responsibilities, obligations, to see if there can be 
19   some structure that we can put into place to incent 
20   market makers and liquidity, perhaps away from maker-
21   taker or as a different market model to maker-taker. 
22             The other ones, we are going to look at the 
23   closing auction process and the two exchanges backing one 
24   another up, NASDQ and NYSE.  More particularly, we have 
25   some concerns about what happens after 3:50.   
0153
 1             So, the -- you know, I think generally the view 
 2   of the subcommittee was the backup plans make a lot of 
 3   sense if there's enough notice to everyone.  But after 
 4   3:50, the -- sort of the VWAP models and so forth have 
 5   raised some concerns about whether we're getting to the 
 6   right price and whether that's the right thing.  So, we 
 7   will spend some time on that. 
 8             And then, the last thing is around capital 
 9   formation, which goes to some of the points that were 
10   raised earlier today, which is, you know, what do we do 
11   for small cap stocks and mid-cap stocks; do we have the 
12   right market models in place to create that kind of 
13   model. 
14             In response to some of the comments this 
15   morning, I will say that we at the subcommittee have, in 
16   fact, asked for issuer representatives regularly.  So, we 
17   have not only asked NASDAQ and NYSE to participate in 
18   these committee meetings, we have gone so far as to ask 
19   for them to suggest issuers who might have an interest in 
20   participating and have not yet had those invitations 
21   taken up. 
22             Thanks to Senator Kaufman, we've had some 
23   people coming in from Weild & Company and other -- and 
24   other areas in the issuer space who have an interest in 
25   this space.  So, we have gotten some feedback in some of 
0154
 1   these areas, but we have not had yet a single issuer who 
 2   has come to a meeting though we would certainly welcome 
 3   them. 
 4             So, with that, I'm happy to take any questions. 
 5    And if anybody else on the subcommittee has any comments 
 6   of anything I missed, please jump in.   
 7             MR. LUPARELLO:  Questions for Eric? 
 8             Rick. 
 9             MR. KETCHUM:  Just on the modable readjustments 
10   of limit up-limit down and the differences of your 
11   committee, I entirely get and agree that the interim day 
12   auctions have not work optimally.  But while you do a 
13   good thing here from the standpoint of extending the time 
14   frame before readjustment on the limit up-limit down, you 
15   could still move down a lot very quickly without the 
16   five-minute period from the standpoint of auctions. 
17             Did you think about whether, as the stock moved 
18   farther down from the standpoint of readjustments to the 
19   limit up-limit down that you'd have longer periods before 
20   you readjust and reset further down so that -- as you 
21   know having been a part of it, the primary reason for 
22   auctions wasn't that anyone thought at the time it was 
23   set that it would work super, but it was a recognition 
24   that almost by definition at some point to suggest that 
25   the electronic trading wasn't providing the alternate 
0155
 1   side without human intervention.  So, there was a desire 
 2   to find some way to have sufficient time for humans to 
 3   evaluate and reset the electronic trading so that the 
 4   other side might come in.   
 5             So, did think about longer periods before 
 6   readjustment as the stock moved farther and farther down? 
 7             MR. NOLL:  We hadn't specifically said, as it 
 8   moved farther and farther away, adjusting the time.  
 9   There was a lot of conversation around what is the 
10   appropriate length of a limit condition.   
11             And I think even though we suggest going to 30 
12   seconds in here, I think that there continues to be a lot 
13   of people on the committee and elsewhere who feel that 
14   that's insufficient as well.  
15             What the correct time should be in the limit 
16   condition, I think, is an open question.  But to your 
17   point, most of the people in the marketplace who would 
18   come back in, you know, said it could be as long as 
19   minutes to be able to sort of reach those conclusions 
20   about how to --  
21             MR. KETCHUM:  Yeah, that would be my concern.  
22   Certainly, as you started talking about 15, 20 percent 
23   moves as opposed to 5. 
24             MR. NOLL:  But again, I think one of the key 
25   points that I want to go back to, which is, nobody loves 
0156
 1   the cascading effects of limit up-limit down.  You know, 
 2   they're not great.  But in fact --  
 3             MR. KETCHUM:  They're better. 
 4             MR. NOLL:  They're better than not having them. 
 5    And I think that's one of the key points that we wanted 
 6   to make sure got expressed. 
 7             MR. LUPARELLO:  Jamil? 
 8             MR. NAZARALI:  If I can just add to what Eric 
 9   said and what we discussed in the subcommittee, we felt 
10   very strongly that market halts are disruptive in the 
11   market.  And they actually have the opposite effect of 
12   what's intended.  So, you know, you have this nice idea 
13   that you have a halt and it'll encourage everyone to take 
14   a pause and enter their orders into the opening 
15   mechanism. 
16             What actually happens is, when the stock gets 
17   halted, most people leave the marketplace.  So, if you're 
18   an electronic algo provider, you might have a lot of 
19   passive liquidity out in multiple marketplaces.  Stock 
20   gets halted; you cancel it.   
21             You might be a market maker like us, and you 
22   have market making all over the place.  Stock gets 
23   halted; you cancel all that liquidity. 
24             And so, while we cared -- you know, we were 
25   open to different time periods for that limit up-limit 
0157
 1   down pause.  We felt really strongly that that opening 
 2   auction should go away and that it also allows you to 
 3   take advantage of the liquidity across the marketplace 
 4   rather than only the liquidity that's entered onto the 
 5   opening auction, which tends to be very thin. 
 6             MR. LUPARELLO:  Joe. 
 7             MR. MECANE:  Just quickly, the only thing I was 
 8   intrigued by or surprised by in the recommendations was 
 9   not conforming the SPX and SPY conventions.  And I think 
10   it said in the memo -- and you just said that was partly 
11   because no one wanted there to have been a market-wide 
12   halt, which I totally understand and agree with. 
13             Is there further discussion around other 
14   potential modifications that should be made to the 
15   market-wide circuit breaker construct that you guys are 
16   thinking about?  Or is it more -- like is it more that 
17   conforming to the futures product is probably the right 
18   decision also, or is more just about not triggering halts 
19   unnecessarily? 
20             MR. NOLL:  Well, I will give my personal bias, 
21   which is I actually am in favor of us conforming to the 
22   futures because I think it's a better real-time way to 
23   determine where the market is or where it's going. 
24             There were some strong objections across the 
25   committee to doing that primarily, I think, driven by the 
0158
 1   fact that there -- this is an equity market and ought to 
 2   be determined by what's going on in the equity space as 
 3   opposed to a derivative space. 
 4             And so, there's -- I think it's an open issue. 
 5   But pretty clearly, I was in the minority with the 
 6   futures position. 
 7             MR. LUPARELLO:  And Eric, I took from your 
 8   comments that you thought that the future was the right 
 9   measurement but perhaps the bands were too narrow.  Is 
10   that fair? 
11             MR.  NOLL:  Yeah.  And I think that's -- I 
12   think that's a -- yeah, I think that's the other sort of 
13   conclusion that people are getting their heads wrapped 
14   around, which is -- I don't necessarily know that the 
15   bands were too narrow.   
16             I think there was a difference between, say, a 
17   noted panic in the marketplace for an exogenous region, 
18   right.  A terrible tragedy happens.  Markets are looking 
19   massively down.  There -- a lot of people are saying -- 
20   yeah, there we think the markets ought to be halted. 
21             When it's just sort of unknown volatility in 
22   the marketplace that just needs a little time to settle 
23   down and shake out, most people were saying -- yeah, then 
24   the market shouldn't halt.  
25             So, it's not just are the bands correctly set 
0159
 1   at their widths, but there was also a -- sort of a 
 2   qualitative sense, which I'm not sure that we could get 
 3   to.  It actually makes me very nervous because who makes 
 4   that call and who actually makes the decision that that's 
 5   a real one and not a real one.   
 6             But there was a real sense that there's a 
 7   qualitative difference to when a halt should be called 
 8   and when it shouldn't be.   
 9             MR. LUPARELLO:  Joe. 
10             MR. RATTERMAN:  Yeah.  I'm not sure I'm going 
11   with this, but I share your concerns that we shouldn't 
12   have halted the market that day, but I think I share that 
13   concern pretty much every day and mostly because we've 
14   never tested it.  It's never actually happened.  We have 
15   -- none of us have any idea how it's going to work. 
16             So, I guess the challenge back to the 
17   subcommittee or to the Commission is, you know, is there 
18   a way somehow that this industry can actually somehow 
19   test under non-volatile conditions how the market reacts 
20   to stopping trading and restarting on a market-wide basis 
21   so that we have more comfort that we've got an actual 
22   safety net underneath of us, because, while we defined it 
23   and on paper it looks great, I don't really know how it's 
24   going to work when we actually hit it. 
25             So, I think we're all scared of it because 
0160
 1   we've never seen it work, and we have no way of testing 
 2   it other than on a Saturday when only about 10 percent of 
 3   the people show up. 
 4             MR. LUPARELLO:  Jamil.   
 5             I'm sorry, Eric.   
 6             MR. NOLL:  No, I was going to say, you know, 
 7   and I think Joe highlighted something that I think is 
 8   resident on the committee, is that we have a bias to have 
 9   the market open as opposed to closed.  And that also 
10   informs our thinking about limit up-limit down as well. 
11             MR. NAZARALI:  Yeah.  And I think another 
12   concern of the committee was that -- and part of the 
13   reason we got this "well, thank God the market wasn't 
14   closed," is we feel like when the markets are halted, 
15   that actually adds to the panic.  And so, the markets 
16   tend to be -- well, they're halted when the market is 
17   down.  That just brings a lot of sellers into the market.  
18             And so, the feeling was, unless it's really an 
19   extreme situation -- and there's lots of international 
20   examples where you can see when the market has been 
21   halted because it goes down.  When it opens, it's 
22   actually opening under even more dire conditions.  So, we 
23   really should think about that and whether nor not that 
24   band should be a little bit wider. 
25             MR. LUPARELLO:  Maureen. 
0161
 1             MS. O'HARA:  I was going to say I thought one 
 2   of the interesting discussions that we had in the 
 3   committee was talking about the fact that part of the 
 4   challenge on August 24 in getting some of the equities 
 5   open is that the futures were all closed.  And with the 
 6   futures closed, the role that the equities had, the 
 7   equity market makers, had in trying to open and some of 
 8   the challenges for the ETFs became clear. 
 9             So, I think, you know, the committee probably 
10   does need to spend a bit more time thinking about these 
11   issues because it is -- you know, it is a challenge.  
12   When the futures is closed at the open, that creates a 
13   series of difficulties that I don't think we have been 
14   able to think through in our discussion that, you know, 
15   markets should open without really making clear exactly 
16   how, I think illustrates that problem. 
17             MR. LUPARELLO:  Reggie? 
18             MR. BROWNE:  I know this was discussed in the 
19   committee.  I don't think there's much to fear with 
20   market-wide halts actually.  And that was -- that's my 
21   opinion.  I believe that the first time it happens is 
22   probably disruptive, but it's an educational environment. 
23    And as education ebbs, I think behavior will be 
24   adjusted.   
25             And so, how do we -- the question is, how do we 
0162
 1   slow down erratic behavior; how do we model behavioral -- 
 2   good behavior in times of stress.  And so, there was a 
 3   lot of discussion on the committee around market-wide 
 4   halts or recognizing market-wide events and then using 
 5   futures as a proxy, you know, for a guidepost. 
 6             But in my view -- this is my personal view -- 
 7   that a market-wide halt over a period of times is a 
 8   behavior adjustment event and not to fear that. 
 9             MR. LUPARELLO:  Just so long as I'm not the 
10   director of Trading and Markets when it happens. 
11             Any other questions for Eric? 
12             Thank you and look forward to further work on 
13   that. 
14             Last but certainly not least, Manisha, we'll -- 
15   can you give us an update on the Customer Issues 
16   Subcommittee?   
17             MS. KIMMEL:  Sure.   
18             So, the primary focus of the Customer Issues 
19   Subcommittee has revolved around four topics.  One is 
20   understanding the retail investor perspective on equity 
21   market structure.  Two is the use of stop orders by 
22   retail investors.  Three is a review of execution quality 
23   statistics from a retail perspective, and fourth is 
24   payment for order flow.  
25             We have met with a group of retail broker 
0163
 1   dealers and reviewed their recommendations, and our 
 2   progress to date is as follows:  first, on retail 
 3   investor perception of equity market structure, we have 
 4   been discussing this issue and the benefits of 
 5   understanding a retail investor perception as it relates 
 6   to equity market structure in a standardized manner on an 
 7   ongoing basis.   
 8             We have met with the Office of Investor 
 9   Advocate, SEC's office, to understand their expected 
10   survey activity.  Additionally, we are in the process of 
11   reviewing academic studies and surveys specifically 
12   focused on investor sentiment. 
13             We expect to complete our analysis over the 
14   next several weeks and make a recommendation on how the 
15   Commission can assess retail investor perception of the 
16   equity markets on a regular basis. 
17             The second issue is the use of stop orders by 
18   retail investors.  And we've discussed the use of stop 
19   orders, particularly in times of volatile conditions.  
20   And while we believe that stop orders, you know, are a 
21   useful tool for investors, they help them mitigate 
22   investment risk in certain circumstances, there are 
23   times, particularly in times of market volatility like 
24   August 24, where prices -- where orders may execute at 
25   prices significantly inferior to market prices. 
0164
 1             As a result, the subcommittee believes that 
 2   firms should review their practices regarding stop 
 3   orders, including firms -- targeted training to 
 4   registered reps; allowing customers to enter stop orders 
 5   directly online with clear and comprehensive disclosures 
 6   prominently provided before the time of order entry; 
 7   additionally looking at other safeguards like potentially 
 8   preventing the use of stop orders without limits or 
 9   restricting the time of day where stop orders may be 
10   executed. 
11             The subcommittee recommends at this point that 
12   the SEC and FINRA consider issuing additional guidance to 
13   further emphasize the importance of these effective 
14   practices relating to stop orders. 
15             The next area we focused on was enhancement and 
16   expansion of execution quality reports, particularly from 
17   a retail perspective.   
18             First off, we talked about execution quality 
19   reports at the individual broker-dealer level.  So, as 
20   you know today, Rule 605 requires market centers, not 
21   broker-dealers, to report.  And we believe that it may be 
22   beneficial to have all broker-dealers, with the possible 
23   exception of small broker-dealers, doing Rule 605 type 
24   reporting. 
25             We recognize that there's an implementation 
0165
 1   cost both initially and ongoing with a compliance burden. 
 2    But having third-party vendors provide this service may 
 3   help mitigate some of these concerns. 
 4             We talked about the fact that there are, you 
 5   know, interest in retail as an individual investor is 
 6   looking at data on -- based on Rule 605 data and that it 
 7   wouldn't be individual investors that would look at 605 
 8   data specifically; it would be the retail broker-dealer 
 9   community looking at that.  But with that data, retail 
10   broker-dealers could create their own simplified versions 
11   of statistics, so rather than trying to dictate what that 
12   simplified version is, allow that to be a retail broker-
13   dealer decision. 
14             Next, we talked about access to execution 
15   quality reports.  We believe that access to Rule 605 and 
16   606 reports should be centralized at the SEC.  We believe 
17   that the existing market structure data and analysis 
18   website could act as a repositor for this data.  Also, 
19   using the SEC's data visualization tool for 605, 606 data 
20   may allow for meaningful comparisons. 
21             We also addressed specific enhancements as it 
22   related to Rule 605 and 606.  And we believe that, in 
23   general, these reports should be both standardized and 
24   meaningful and reflect some of the changes to the market 
25   since they were first introduced. 
0166
 1             I mentioned we took a retail perspective, which 
 2   included discussion with retail broker-dealers and a 
 3   review of issues raised in comment letters.  So, we 
 4   believe that there is opportunity for enhancements in 
 5   both 605 and 606. 
 6             Just to give you a sample of some of those for 
 7   -- with respect to 605, the inclusion of odd lots; the 
 8   segregation of immediate or cancel orders; the inclusion 
 9   of quoted spread and enhanced liquidity; and the revision 
10   of the frequently asked questions to assign an NBBO for 
11   orders received in the same second.  So, currently, the 
12   FAQs say you can have a neutral algorithm.  We believe it 
13   should be specific. 
14             With FINRA's recently approved proposal to get 
15   clock-synced to 50 milliseconds and CAT coming up with 
16   the millisecond timestamps, there's an opportunity to do 
17   things to the millisecond prior to the order being 
18   received. 
19             With respect to Rule 606, we've talked about 
20   dividing the data, instead of by listing market, to 
21   consider division between -- by S&P 500 securities and 
22   other NMS securities; to include a new section that gives 
23   OTC equity data; the segregation of data currently in the 
24   other category -- take some of that other category date 
25   out and put it into its own sections.  Those sections 
0167
 1   could include market open/close orders, stop, stop on 
 2   quote orders, and then have odd lots be their own row, if 
 3   you will. 
 4             One of the other things we discussed is the 
 5   consistent identification of routing destination across 
 6   Rule 606 reporters.  That becomes especially important if 
 7   you put this data into a tool so that you could make 
 8   meaningful comparisons. 
 9             The inclusion of plain English descriptions, 
10   definitions, and facts -- and, you know, we'd love to see 
11   this tied with the recommendation from Trading Venue 
12   regulations, which would say -- as we were to implement 
13   these rules, these facts would be created simultaneously 
14   to the release of the rule. 
15             While we recognize that Rule 606 -- even though 
16   it's easier to look at than 605 data, it still may not be 
17   used by retail investors.  We think the move to 
18   simplified language and plain English definitions would 
19   force consistency and standardization that we may not 
20   have today. 
21             We also discussed inclusion of execution 
22   quality statistics by routing destination. 
23             The fourth topic we covered was payment for 
24   order flow.  The subcommittee has not come to a consensus 
25   on action with respect to payment for order flow.  Our 
0168
 1   discussions have ranged from prohibiting the use of 
 2   payment for order flow, managing conflicts of interest 
 3   and opportunities for enhancing existing or requiring new 
 4   disclosures.   
 5             The next steps of this group are to discuss 
 6   institutional order routing and execution quality.  We 
 7   agree with the conversations earlier today that retail 
 8   investors are also represented by institutional order 
 9   flow. 
10             We had previously tabled our discussion of 
11   institutional execution quality, given our anticipation 
12   of SEC rulemaking on institutional order routing and 
13   execution-quality statistics.  So, we're hopeful that we 
14   would be able to review that rulemaking and also talk to 
15   institutional-focused industry groups to better 
16   understand and make a final recommendation that also 
17   includes institutional order routing and execution-
18   quality use of 605/606. 
19             Additionally, we plan on meeting with the co-
20   chairs of the Market Structure Subcommittee of the SEC 
21   Investor Advisory Committee and other industry groups. 
22             I welcome any questions or comments. 
23             MR. LUPARELLO:  Questions for Manisha? 
24             That's an exhaustive list.  And I would remind 
25   folks that on the two very helpful and very thorough 
0169
 1   subcommittee reports that all these recommendations, all 
 2   these topics for conversation will get repeated full air 
 3   time at future committee meetings.  And again -- so, 
 4   slightly unfortunate that we didn't have the capacity to 
 5   deal with all of them at once.  But to give them the time 
 6   and energy they deserve, I think it's best to hold them 
 7   to our next committee meeting. 
 8             Unless I hear any other topics, any other 
 9   questions, I will entertain a motion to adjourn.  The 
10   Chair has already sidled away, so assume that that's a 
11   good sign.   
12             Yes, exactly.  That is automatically both a 
13   motion and a second.  But if I could officially have a 
14   motion and a second, that would be appreciated. 
15             MS. KIMMEL:  Second. 
16             A PARTICIPANT:  I --  
17             MR. LUPARELLO:  I have many of both.   
18             Thank you all again for your extraordinarily 
19   hard work and your thought on these very important 
20   issues. 
21             And we are adjourned. 
22             (Whereupon, at 2:36 p.m., the meeting was 
23   adjourned.) 
24              
25    
0170
 1                   PROOFREADER'S CERTIFICATE 
 2    
 3   In The Matter of:   EQUITY MARKET STRUCTURE 
 4                       ADVISORY COMMITTEE MEETING     
 5   File Number:        OS-0426 
 6   Date:               April 26, 2016 
 7   Location:           Washington, D.C. 
 8    
 9             This is to certify that I, Nicholas Wagner, 
10   (the undersigned), do hereby swear and affirm that the 
11   attached proceedings before the U.S. Securities and 
12   Exchange Commission were held according to the record and 
13   that this is the original, complete, true and accurate 
14   transcript that has been compared to the reporting or 
15   recording accomplished at the hearing. 
16    
17   _______________________       _______________________ 
18   (Proofreader's Name)          (Date) 
19    
20    
21    
22    
23    
24    
25