0001
 1         UNITED STATES SECURITIES AND EXCHANGE COMMISSION
 2   
 3   
 4   
 5   
 6                   ROUNDTABLE ON DECIMALIZATION
 7   
 8   
 9   
10   
11   
12                     TUESDAY, FEBRUARY 5, 2013
13                     (Amended:  March 13, 2013)
14   
15   
16   
17   
18   
19   
20   
21   
22   
23           U.S. Securities and Exchange Commission 
24                      100 F Street, N.E. 
25                       Washington, D.C. 
0002
 1   SEC COMMISSIONERS PRESENT:
 2        Elisse Walter, Chairman
 3        Daniel Gallagher, Commissioner
 4   
 5   PARTICIPANTS PRESENT:
 6        James Burns
 7        Colin Clark
 8        Chris Concannon
 9        Brian B. Conroy
10        Kevin Cronin
11        R. Cromwell Coulson
12        Joshua L. Green
13        Frank M. Hatheway
14        Chris Isaacson
15        Paul Jiganti
16        Scott Kupor
17        Craig Lewis
18        Maureen McCarthy
19        Lona Nallengara
20        Maureen O'Hara
21        Adam V. Reed
22        Stephen Sachs
23        Jeffrey M. Solomon
24        David Weild
25        Kent Womack
0003
 1                          C O N T E N T S
 2                                                            PAGE
 3   Opening Statement
 4        James Burns                                          4
 5        Daniel Gallagher, Commissioner, SEC                  5   
 6   
 7   Panel 1:  Evaluating Concerns Relating to Tick
 8             Size for Small and Middle Capitalization
 9             Companies                                      31
10   
11   Panel 2:  Evaluating Concerns Relating to Tick
12             Size for the Securities Market 
13             Generally                                      87
14   
15   Panel 3:  Studying the Effects of Alternative
16             Tick Sizes                                    146
17   
18   
19   
20   
21   
22   
23   
24   
25   
0004
 1                P R O C E E D I N G S  (9:33 a.m.)
 2             MR. BURNS:  I think we will try to get started. 
 3   Good morning, everyone.  I'm Jim Burns, Deputy Director of
 4   the Division of Trading and Markets at the SEC.  
 5             It's my pleasure to welcome you to the Commission's
 6   Roundtable on Decimalization.  We are pleased you have been
 7   able to join us for what we anticipate will be an informative
 8   and productive series of discussions.
 9             We had hoped at this point to introduce our
10   Chairman, Elisse Walter, and our Acting Director, John
11   Ramsay, who are in Brussels.  They will be joining us by live
12   feed in a little bit.  We will turn to them when I get the
13   cue they are in situ.
14             While we are waiting for them, I would like to
15   convey the regards of some of the Commissioners who are not
16   able to join us today.  Commissioners Paredes and Aguilar are
17   not able to attend in person but they send their regards and
18   Commissioner Aguilar in particular sends his regrets that
19   illness prevents him from being here in person.  I understand
20   he is following us by webcast.
21             We are very pleased to welcome Commissioner
22   Gallagher who joins us here on the podium.  Commissioner,
23   thank you for being with us.  We would appreciate any
24   comments or observations you might wish to share here at the
25   outset or as the morning progresses.
0005
 1             COMMISSIONER GALLAGHER:  It's good to be here.  I
 2   was going to make a little joke about the Commissioners who
 3   aren't here, but I know Luis is watching, and I know he'll
 4   call me in his hoarse voice and yell at me if I make a bad
 5   joke, so I won't do it.
 6             It's great to see everybody here.  I know there are
 7   a lot of people watching on the web, too.  
 8             I think this Roundtable is just a great indication
 9   of how important what might seem like small issues are to the
10   work of the Commission and to investors.
11             Something like decimalization, given our agenda
12   these days with the Dodd-Frank rulemaking's, with the JOBS
13   Act, and with our day to day work, could just be overlooked
14   as something unimportant.
15             To the contrary, it's a very important issue as
16   Congress told us when they mandated the study that the staff
17   carried out last year in the JOBS Act.
18             I am very agnostic as to where this goes, open
19   minded.  This is truly a Roundtable where I think we have the
20   right people here.  We have the right learning and expertise. 
21   The Commission as well as the Commission staff can learn and
22   make a decision as to how to proceed.
23             As we saw last Friday with our Small Business
24   Committee and their recommendation that we move down the path
25   towards a pilot in this space.  This is a real issue that
0006
 1   impacts capital formation at the very important, critical
 2   early stages for companies, I think, considering IPOs.
 3             I think because of that, we should take it
 4   incredibly seriously.  I'm looking forward to the
 5   interactions today.  I will, as Jim pointed out, jump in
 6   where I see fit.
 7             Thanks everybody for being here.  Thanks to the
 8   staff for putting it together.  Great job.
 9             MR. BURNS:  Thank you, Commissioner Gallagher. 
10   Thank you, too, for your leadership and dedication to the
11   staff's efforts on this and other important initiatives.
12             For starters, the staff has had a hard time
13   wheedling down the list of invitee's to the esteemed
14   panelists you have here today.
15             I want to remind folks at the outset, other
16   interested parties, please offer your comments or
17   observations.  We would like to include those in the comment
18   file.  Already, we have a very robust comment file.  We do
19   appreciate your comments as they may come in.
20             A special thanks -- we will do this periodically
21   throughout -- to the panelists before us and on the panels
22   that follow for the time and expertise they are lending to
23   the Commission and the staff as we assess a topic of great
24   importance.
25             Quickly, in particular, today's over-arching topic
0007
 1   is whether the current tick size regime in the U.S. equities
 2   markets can be enhanced to more fully promote vital public
 3   policy objectives, such as fair and efficient trading, the
 4   protection of investors, and vigorous capital formation.
 5             We are tackling this in three parts today.  The
 6   first panel is focused on tick size as they affect small and
 7   mid-cap companies.  The second panel is focused on tick sizes
 8   in the context of the equities markets more generally.
 9             After lunch, the third panel will focus on ways to
10   generate useful data to enable the Commission to evaluate
11   tick size issues in as disciplined and a responsible manner
12   as possible.
13             Among other things, the third panel will assess
14   whether a pilot program might be implemented to assign
15   varying tick sizes to a control group of stocks of different
16   types of companies.
17             The objective of such a pilot as well as other
18   potential means of generating data would be to establish a
19   solid empirical basis for any action the Commission might
20   determine to take or not take in the future.
21             We are optimistic that a data driven approach can
22   be developed that will prove fruitful for addressing tick
23   sizes, and more broadly, I think this type of data driven
24   approach -- we all think, I believe -- can provide an useful
25   template for addressing many of the other complex and
0008
 1   pressing market structure issues that currently face the
 2   Commission.
 3             I can run down a number of important steps we have
 4   taken in recent years, market-wide circuit breakers, market
 5   access rules, updated SRO rules relating to breaking of
 6   trades, sub-quotes, or other enhancements on the horizon or
 7   with us now, audit trail, large trader, the new MIDAS system
 8   that will collect and facilitate data analysis within the
 9   Division and among the offices of the Agency, and the quantum
10   leap in hiring of personnel with specialized expertise.
11             These initiatives have begun and should continue to
12   strengthen our approach to the U.S. equity market structure. 
13   There is plenty to do.
14             In addition to the tick size issues, there are many
15   other issues relating to market structure that we should
16   further examine and we will endeavor to do so.
17             As many of you know, our concept release asked
18   questions about tick size a number of years ago.  There are
19   many other areas of endeavor we want to pursue under that
20   concept release.
21             Whatever the Commission may do, the most promising
22   path forward is to adopt a data driven approach to these
23   issues, to provide a solid empirical basis for the
24   Commission, to enable it to address open issues and determine
25   what if any regulatory actions are appropriate and how they
0009
 1   might best be calibrated.
 2             In sum, we hope and expect the Commission will
 3   benefit very much in the coming months from various sources
 4   of empirical research. 
 5             Today's discussion on tick sizes is an excellent
 6   example of how we can pursue an approach that enables
 7   progress on a wide range of other issues.
 8             I am not sure if we have the Chairman yet.  If not,
 9   I'd like to turn it to a key leader in our efforts, Craig
10   Lewis, the Commission's Chief Economist, and Director of
11   Risk, Division of Risk, Strategy and Financial Innovation,
12   for his introductory remarks.
13             Craig?
14             MR. LEWIS:  Thank you, Jim.  I'd like also to begin
15   by thanking everyone for joining us today for this important
16   Roundtable to discuss the impact of tick size on small and
17   mid-sized companies, market professionals, investors and U.S.
18   security markets.
19             As always, I must make clear that the views I
20   express today are my own and do not necessarily reflect the
21   views of the Commission or my colleagues on the Commission
22   staff.
23             The genesis of today's Roundtable is a study
24   conducted by staff in my Division, the Division of Risk,
25   Strategy and Financial Innovation, in response to a
0010
 1   congressional mandate included in the JOBS Act.
 2             Specifically, the JOBS Act required the Commission
 3   to examine the effects of decimalization on initial public
 4   offering's and small and middle capitalization companies.
 5             Staff in RSFI worked diligently with staff from the
 6   Division of Trading and Markets and the Division of
 7   Corporation Finance to complete this study within the 90 time
 8   frame mandated by Congress.
 9             We took a three pronged approach to conducting the
10   study within this time frame.  Specifically, the study
11   entailed (1) reviewing empirical studies regarding tick size
12   and decimalization; (2) participation in a meeting of the SEC
13   Advisory Committee on Small and Emerging Companies on the
14   impact of market structure on small and middle capitalization
15   companies, and (3) a survey of the tick size conventions in
16   non-U.S. markets.
17             A report of our study was delivered to Congress in
18   July of last year.  The report, which is available on the
19   Commission's website, explained that the impact of mandating
20   an increase in the minimum tick size for small capitalization
21   companies on the structure of our markets and on the
22   willingness of small companies to undertake initial public
23   offering's, is at best uncertain.
24             In light of this conclusion, the report ultimately
25   recommended that the Commission not proceed with specific
0011
 1   rulemaking to increase tick sizes but should consider
 2   additional steps that may be needed to determine whether
 3   rulemaking should be undertaken in the future.
 4             One of the additional steps recommended in the
 5   report was for the Commission to solicit the views of
 6   investors, companies, market professionals, academic's and
 7   other interested parties on the topic of tick sizes, how to
 8   best study its effects on IPOs, trading and liquidity for
 9   small and middle capitalization companies, and what if any
10   changes should be considered.
11             Indeed, this recommendation is what brings us here
12   today.  
13             Before we start soliciting your views, I thought it
14   may be helpful to discuss examples of what the Commission
15   staff's findings have revealed thus far.
16             Although the staff ultimately concluded that the
17   impact of mandating an increase in the minimum tick size is
18   still uncertain, the staff did report findings that should
19   help illuminate our discussions today.
20             First, the staff undertook a literature review of a
21   number of academic studies that discussed the effects of the
22   change to decimal pricing on overall market quality and where
23   possible, the impact on small and middle capitalization
24   stocks.
25             As summarized in this report, the academic
0012
 1   literature documents not only lower spreads after
 2   decimalization but also smaller trade sizes, decreased long
 3   run volatility, and no apparent reduction in market maker
 4   profitability.
 5             On net, these findings suggest that the initial
 6   effect of decimalization was improved market quality.  Most
 7   notably, for large capitalization securities.  
 8             For those papers that examined smaller and middle
 9   capitalization securities, the effect, particularly for
10   spreads, appears to be relatively minor, but for reasons
11   outlined in the report, it is difficult to draw strong
12   conclusions from a review of these available studies with
13   respect to potential adverse effects on the change to
14   decimals on small and middle capitalization companies today.
15             The staff also considered the discussion and
16   material presented at a meeting of the SEC Advisory Committee
17   on Small and Emerging Companies on market structure issues.
18             As the staff explained, at that time, members of
19   the Advisory Committee commented that it may be hard to
20   disentangle the impact of decimalization on small company
21   IPOs for several concurrent reasons, including the enactment
22   of the Sarbanes-Oxley Act in 2002, and the more recent
23   emergence of high frequency trading in dark pools.
24             I would note, however, that just last week, the
25   Advisory Committee, after further review, recommended that
0013
 1   the Commission adopt rules to increase tick size for smaller
 2   Exchange listed companies in the United States that will
 3   allow such companies to choose their own tick size within a
 4   range designated by the Commission.
 5             The staff study also surveyed tick sizes in a
 6   variety of countries that have significant smaller company
 7   IPO activity.  This review revealed that other countries use
 8   multiple tick sizes in contract to the one size fits all
 9   approach in the United States, which suggests there may be
10   viable and perhaps preferable alternatives to uniform tick
11   size rules. 
12             Although it also is worth noting that the study
13   explains that regulatory decimalization in the market lowered
14   the minimum allowable tick size, it did not mandate market
15   participants quote narrower spreads and quoting narrower
16   spreads appears to have been the result of continued market
17   forces.
18             Thus, although we are left with many questions, our
19   study has established a foundation for analyzing issues
20   related to tick sizes.
21             This Roundtable represents another step in
22   collecting valuable information on how best to proceed in
23   this complex area of regulation, and will hopefully give us a
24   better understanding of the costs and benefits of potentially
25   changing tick sizes.
0014
 1             Finally, I would be remiss if I failed to echo
 2   Jim's sentiment that the study also provides an example of
 3   how data driven empirical analysis can illuminate policy
 4   discussions at the Commission.
 5             I wholeheartedly believe that this approach if done
 6   appropriately strengthens the Commission's policy
 7   considerations.
 8             My Division hopes to continue to work as partners
 9   with our colleagues in the other divisions to develop similar
10   approaches to addressing the numerous regulatory issues that
11   face the Commission, including the market structure projects
12   that Jim mentioned.
13             I want to thank you again for attending this
14   important Roundtable.  I look forward to listening to your
15   views and the helpful and productive discussions that we will
16   have today.
17             Now, I will turn the floor back over to Jim for
18   further comments.
19             MR. BURNS:  Thank you very much, Craig.  I thank
20   you on behalf of Trading and Markets for all the efforts you
21   and your staff have put into this and some of the other
22   important equity market structure initiatives.  Indeed, many
23   other initiatives we take on.
24             I am not sure whether we have the Chairman with us
25   yet.  
0015
 1             I am pleased to introduce Lona Nallengara, the
 2   Acting Director in the Division of Corporation Finance, who
 3   along with his Corporation Finance colleagues, leads many
 4   staff efforts relating to issues that affect small and mid-
 5   cap companies.
 6             Lona, we will give you a signal when the Chairman
 7   and John are on line.
 8             MR. NALLENGARA:  Okay.  We are going to start the
 9   first panel?
10             MR. BURNS:  If you folks can indulge us, we will
11   let you know if we need to pause for a moment.  We will tack
12   on some time at the end.
13             MR. NALLENGARA:  Thank you, Jim.  Our first panel
14   is titled "Evaluating Concerns Relating to Tick Size for
15   Small and Middle Capitalization Companies."
16             Before we get started, I thought I'd introduce some
17   of the staff members we have today that are going to help
18   facilitate the discussion.  
19             You have already met Jim and Craig.  To my right,
20   Kathleen Hanley, is our Deputy Director in the Division of
21   Risk, Strategy and Financial Innovation, and also our Deputy
22   Chief Economist.
23             Down at the other side working from the end, Daniel
24   Gray, Heather Seidel, David Shillman, and Gregg Berman.  They
25   are in the Trading and Markets Division, and Amy Edwards,
0016
 1   also in Risk, Strategy and Financial Innovation.
 2             Together, this group and their teams are leading
 3   our efforts in thinking and re-thinking market structure
 4   issues.  We are glad they are here today to help us with the
 5   discussion.
 6             I would also like to welcome our panelists today. 
 7   We have an impressive group here today.  We appreciate the
 8   time they have taken to join us.
 9             I would like to ask them to each introduce him or
10   herself to everyone here.  We are hoping to have as much time
11   as possible for both the discussion as well as the Chairman's
12   remarks, so if I can ask each of the panelists to keep their
13   introductions brief.  There is a lot of information about
14   them in the materials and on our website.
15             I will start with -- as a product of the Canadian
16   Educational System -- Professor Womack, to start with the
17   introductions.
18             MR. WOMACK:  Thank you very much.  I have a career
19   that started in investment banking.  After that, I went back
20   to get a Ph.D. at Cornell.  My research has been about the
21   equity markets, specifically IPO markets, conflicts of
22   interest and research on IPO firms.
23             When we get to the topic today of how this might
24   impact research, the research environment for securities,
25   hopefully I'll be able to add a couple of things.
0017
 1             Right now, I'm at the University of Toronto in
 2   Canada, and expect to stay there for my career.
 3             MR. BURNS:  I'm going to break in right here.  Do
 4   we have the Chairman and John Ramsay?
 5             CHAIRMAN WALTER:  You do.
 6             MR. BURNS:  Terrific.  We are delighted you could
 7   join us.  We are in the middle of introductions, so we will
 8   turn the floor over to you and then to John, and then we will
 9   continue with introductions. Thanks for joining us.
10             CHAIRMAN WALTER:  Thank you very much, Jim, and
11   after the drama it has taken to get us here and actually on
12   camera, it would be nice if we had something very dramatic to
13   say, but I can assure that, that's not the case.
14             MR. RAMSAY:  We don't.
15             CHAIRMAN WALTER:  Good morning to everyone.  We are
16   sorry to interrupt the flow but we did want to be able to say
17   a few words to you.  Welcome, all of you, and to say that
18   John and I both regret that we are unable to join you.
19             John and I are both out of the country today
20   discussing over-the-counter derivatives with our counterparts
21   from Europe, Asia, and the Americas, and thus, not able to
22   join you for this important event.
23             We do, however, want to let you know that we are
24   very interested in the issues concerning decimalization, and
25   we are very eager to hear from all of you and to get the
0018
 1   results of your discussions today.
 2             Promoting capital formation and economic growth
 3   are, as you know, central objectives of the SEC, and the
 4   equity markets we regulate, and we have a special concern for
 5   the needs of smaller companies that need access to capital to
 6   grow, expand, and flourish.
 7             I know that many feel that today's market
 8   structure, including the current tick size regime, does not
 9   recognize the particular needs of smaller companies, and that
10   a detailed examination of the current structure is important
11   and appropriate.
12             I am pleased that this issue is getting needed
13   attention today, and that today's panels include a broad
14   range of participants who will be able to speak to the needs
15   of smaller companies and give the current structure the
16   scrutiny that it deserves.
17             We want to thank all of you for your participation
18   and for lending your expertise to this effort.
19             I am also very interested in discussion today about
20   the potential for a pilot program and how it might be formed
21   to put to the test some of the ideas being discussed today.  
22             Given the complex nature of our modern markets, we
23   have increasingly called upon the ability to test potentially
24   significant structural changes in a well thought out and
25   controlled fashion that can provide us all with the types of
0019
 1   data and information we need to make informed decisions.
 2             It will be interesting to hear from all of you, if
 3   you think that would be a worthwhile endeavor here, and if
 4   so, how you would structure it.
 5             As I said, I'm grateful to all of you for your
 6   willingness to share your time and expertise, and I'm also
 7   grateful that Craig, Lona, Jim, and their staffs, as well as
 8   Commissioner Gallagher, will be participating in our absence.
 9             You have a dedicated and insightful staff team, and
10   they will be briefing me fully on today's discussions and how
11   we can best move forward in the future.
12             Thank you, and I'll turn the floor over to the
13   Director of our Division of Trading and Markets, John Ramsay.
14             MR. RAMSAY:  Thank you, Chairman Walter.  I know
15   the staff of the Division of Trading and Markets and other
16   Divisions have spent a great deal of time and effort in
17   planning this event.
18             From my Division in particular, let me especially
19   thank Deputy Director Jim Burns, Gregg Berman, the new
20   Associate Director for our new Office of Analytics and
21   Research.  Heather Seidel, Dave Shillman, Dan Gray, Ilya
22   Fradkin, Steve Kuan, and Kelly Riley.
23             Thanks also to Commissioner Gallagher and to the
24   panelists who have given their valuable time to contribute to
25   the discussion.  
0020
 1             I hope you will take this effort as an indication
 2   of how seriously we are considering the funding needs of
 3   smaller companies and the topic of decimalization in
 4   particular.
 5             I know this event will do much to advance the
 6   discussion around this multi-faceted topic, but beyond
 7   advancing the dialogue, I also hope that the insight gleaned
 8   today will help the staff to develop a concrete course of
 9   action, pilot program or otherwise, going forward, and beyond
10   this issue, I hope that this effort may serve as a model as
11   we review other aspects of our equity market structure.
12             What that means for me is being willing to
13   challenge old assumptions and think creatively, and where
14   possible, gather hard data that can inform our policy
15   choices.
16             This has to be a group effort, the staff working
17   with every constituency that is affected by a particular
18   issue, and if done in that way, is sure to pay off for 
19   investors and also the companies that rely on them to prosper.
20             Thanks again for all of you being there.  Again, to
21   the panelists, for taking their very valuable time.
22             (Live feed connection lost.)
23             MR. BURNS:  I think he was going to say thank you.
24             (Laughter.)
25             MR. BURNS:  Just like the Super Bowl power going
0021
 1   out.
 2             David almost doesn't need an introduction here, but
 3   David?
 4             MR. WEILD:  Thank you.  First of all, I want to say
 5   thank you for including me, and I want to just acknowledge
 6   the fact that I think you have some terrific public servants
 7   here.  I have enjoyed the opportunity to get to know people
 8   at the SEC.  I think it's a tough spot to be, but I think you
 9   are doing a terrific job, and with a little bit of luck, will
10   make some progress.
11             I was former Vice Chairman of the NASDAQ Stock
12   Market.  I oversaw it at the time, more than 4,000 listed
13   companies. I followed my mentor, Hardwick Simmons, over
14   there.
15             Prior to that, I had run Corporate Finance and I
16   ran Equity Capital Markets, pricing over 1,000 equity
17   transactions, about 500 IPOs, during the course of my career.
18             For a while, I was the co-Chair of Strategic
19   Planning for the Investment Bank, Equity Research Sales and
20   Trading, obviously prior to the Global Research Analyst
21   settlement back when you could do it that way.
22             At Grant Thornton, who has been absolutely terrific
23   supporting our work, we have written a number of sort of
24   thought leadership pieces, which I think informed much of the
25   JOBS Act, including the first one in 2008, IPOs in the ICU.
0022
 1             That looked at the very material drop off in the
 2   number of IPOs, particularly small IPOs, that started to
 3   occur in the late 1990s, coincident with what I call part of
 4   the trinity here, which is the order handling rules in 1997,
 5   followed by Regulation ATS, alternative trading systems,
 6   culminating with decimalization in 2001.
 7             Then we followed that up with a paper called "The
 8   Wake Up Call for America," which actually started to document
 9   the systemic decline in the number of listed companies in the
10   United States, the punch line of which is we have lost 44
11   percent of all listed companies since the peak in 1997.  
12        Think about that number.  It's sort of staggering.  This
13   is operating companies.  We take out the Funds.  We are down
14   44 percent.    
15             For me, what I'm passionate about, is how do we
16   reverse that trend because we think it's killing jobs in this
17   country, we think we have lost about 10 million jobs in terms
18   of opportunity costs.  We can debate about that.
19             We think it's a big number particularly when you
20   look at Enrico Moretti's work, Professor at the University at
21   Berkeley, that says there are five service sector jobs that
22   are created for every one technology job.
23             Thank you.
24             MR. SOLOMON:  Hi.  My name is Jeff Solomon.  I'm
25   actually the CEO of a growth investment bank called Cowen and
0023
 1   Company.  We do focus on small company capital formation in
 2   both equity and the debt markets.
 3             I'm also a member of the Committee on Capital
 4   Markets Regulation, and really today, an unofficial
 5   Ambassador of a number of equity market participants, buy
 6   side/sell side clients and non-clients alike, with whom I've
 7   engaged on this topic.
 8             I also was a huge proponent of the decimalization
 9   move at the end of the last decade or the beginning of the
10   last decade, and I feel like I've come almost full circle on
11   it, particularly as it relates to small companies.
12             Widening tick increments for small cap stocks I
13   definitely think will help to foster a better ecosystem, and
14   it brings back fundamental investors who are otherwise
15   paralyzed by the speed of the market and the lack of real
16   fundamental buyers and sellers at finite gathering points.
17             My view is that widening increments will actually
18   enable firms like mine to begin initiating more research on
19   small cap companies, once again, which in turn spawns the
20   necessary liquidity in the marketplace for new issuers to
21   consider doing IPOs, IPOs that will raise capital to hire
22   more people, as David mentioned.
23             I read David's piece.  It certainly motivated me to
24   try to do something more than just talk about it.  It
25   inspired me to get involved, to see if we could make some
0024
 1   positive change.
 2             While I have a definitive view that decimalization
 3   is one of the root causes of the current equity market
 4   dysfunction for small caps, I'm also willing to allow that
 5   there are other factors contributing to the lack of
 6   liquidity.
 7             I'm not polemic about that.  What I would say is
 8   I'm here primarily to urge you to consider trying a pilot
 9   program.  At the end of the day, we will know whether it
10   works or doesn't work only after we have seen it.  
11             There will be a number of people on this panel and
12   maybe subsequent panels who will give you tons of views and
13   tons of studies that will support or rebut or claim, and I
14   think the answer is we just don't know until we try.
15             That's the purpose of the pilot program, which is
16   to really to try to let us see if it does get better, and
17   let's give it enough time to develop.
18             As we have our discussions here today, I found an
19   interesting quote from Professor John Kay who actually did a
20   similar kind of study in the U.K., talking about their equity
21   markets.
22             He did testimony last year at the National
23   Association of Pension Funds in December and he said the
24   following:  "If equity markets provide the link between the
25   activities of publicly traded companies and returns to
0025
 1   beneficiaries, it follows that the functions of equity
 2   markets are to enhance the capabilities of publicly traded
 3   companies on the one hand and to achieve the best possible
 4   returns to saver's on the other."
 5             This, he said, "Is the central criteria by which
 6   the performance of equity markets is to be judged."
 7             Hopefully, that provides some context, and we are
 8   here to help in any way we can.
 9             MR. NALLENGARA:  Thank you.  Ms. O'Hara?
10             MS. O'HARA:  Good morning.  I'm Maureen O'Hara. 
11   I'm the Robert Purcell Professor of Finance at Cornell.  I've
12   been there forever.  I do research on a wide range of issues
13   in market micro-structure.  I've worked on a wide range of
14   topics including the effects of fragmentation on market
15   quality, the effects of high frequency trading on how market
16   making works.
17             Lately, I've been working on issues relating to
18   liquidity or illiquidity of stocks, particularly those with
19   low prices.
20             I've also in the past worked on the liquidity of
21   IPOs, and in particular, what happens to IPOs as they enter
22   into the market with respect to the trading of IPOs, the
23   roles of the underwriter market maker and the roles of
24   others.
25             Certainly, a wide range of issues that come to the
0026
 1   floor today.  I think a lot of these issues that we are going
 2   to be speaking about cover a wide range of issues, including
 3   sort of the liquidity costs that individual investors face.
 4             I'm looking forward to hearing from my colleagues. 
 5   I also, in addition to my day job, am Chairman of the Board
 6   of Investment Technology Group, ITG.  We do a lot of the
 7   transaction costs analysis in the market.
 8             Through our studies we have a feel for how
 9   transaction costs have changed in these markets, and I am
10   looking forward to hearing the comments from my colleagues
11   today.
12             MR. KUPOR:  Good morning.  My name is Scott Kupor. 
13   I'm the Managing Partner at Andreessen Horowitz.  We are a
14   three and a half year old venture capital firm focused
15   principally on investing in IT companies.  We do invest
16   multi-stage, so anywhere from very early stage companies all
17   the way to growth capital opportunities.
18             We also have the privilege of all of us having come
19   from the operating world, so we have had some experience both
20   financing new companies as well as actually being part of
21   companies that have gone through the IPO process.
22             I'm very excited today to be a part of the panel
23   and hopefully lend some perspectives based on that
24   experience.
25             Thank you.
0027
 1             MR. GREEN:  My name is Josh Green.  Thanks very
 2   much for inviting me to be here today.  I began in Silicon
 3   Valley in the 1980s as a corporate and securities lawyer.  I
 4   had the privilege of working with entrepreneurs on taking
 5   over 100 of their companies public and saw the rise of the
 6   four horsemen, the demise of the four horsemen, and have been
 7   doing this for quite a long time.
 8             In 2006, I decided to leave the legal profession
 9   and join Mohr Davidow as a general partner investing in clean
10   technology investments.
11             I am today the Chairman Elect of the National
12   Venture Capital Association.  We represent about 90 percent
13   of the overall venture capital industry.
14             As I think most of you know, the NVCA was a leader
15   in the adoption of the JOBS Act, specifically under Title I
16   of the JOBS Act.  We believe it is very effective -- although
17   it's quite preliminary at this point -- on ramping of
18   companies to the public markets.
19             The issue now before us is the public markets are
20   the highway and they have potholes.  One of those potholes
21   has to do with decimalization.
22             I look forward to sharing my thoughts on how we can
23   improve the system through pilot programs to derive empirical
24   data that will ultimately encourage more capital formation by
25   small and emerging companies.
0028
 1             MR. COULSON:  Hi.  I'm Cromwell Coulson.  I am the
 2   CEO of OTC Markets Group.  We operate financial marketplaces
 3   for 10,000 U.S. and global securities.  Our trading system is
 4   an SEC registered alternative trading system.
 5             We have a broad range of securities, 1,500 ADRs,
 6   1,500 ordinary shares, 600 small community banks across the
 7   country.  We have 3,000 SEC reporting companies.  We have 500
 8   or 600 non-SEC reporting companies that use our disclosure
 9   services to publish information into the market.
10             We have a mix, all from the very high range global
11   500s down to the smallest penny stocks.  We have very strong
12   less active, less liquid securities.
13             In the search for an efficient market, we may have
14   made our markets too slippery for investors.  I think the
15   most important thing is how do we make the investor
16   experience when they are buying or selling us small
17   securities that they get liquidity, that they see liquidity,
18   that they don't see a tiny amount of liquidity that then when
19   they take it, yo-yo's away before they can buy the amount
20   they want to own.
21             That's what we have done.  We have created a market
22   where it's very slippery as we have gone to remove friction.
23             We have experience with price increments because
24   back in 2004, we had Professor Christy take a look at our
25   trading data through that year.
0029
 1             He came up with the idea that marketplaces should
 2   organize quoting activity so that liquidity would cluster. 
 3   He found natural clusters in price increments in our
 4   marketplace.  
 5             We implemented those.  Those actually worked really
 6   well because our goal was to increase displayed liquidity by
 7   broker-dealers.
 8             However, as manning and limit order protection came
 9   along, FINRA did not share our view that orders should also
10   have increments, and you can't really have "increments"
11   without order increments because there is a lot of reverse
12   manning problems, which is technical speak for a broker
13   buying at one price and having to sell at a worse price
14   because it's not far enough away from the bid.
15             What we did learn was you need to have both quotes
16   and orders.  Two, we recently had the tier sizes of our
17   marketplaces lowered by FINRA rules.
18             We have seen a lot of interesting data there, too. 
19   We lost about 15 percent of our broker-dealer's price quotes. 
20   They didn't leave the market.  They went un-priced.  
21             We actually didn't see a contraction of spreads. 
22   They stayed the same in community banks, everything else
23   widened a little bit.  We saw more inter-day volatility.  We
24   saw significantly less, over half of our displayed liquidity
25   went away because the tier sizes were lowered.
0030
 1             This is what's really great about actually making
 2   changes and watching it.  As we were going through the debate
 3   with FINRA, their reviews were saying broker-dealers aren't
 4   going to lower their tier sizes to the minimum if we lowered
 5   the minimum.  You can't be sure of that.  Well, we can be
 6   now.
 7             There was also a viewpoint that if the minimum tier
 8   size was lowered, there would be more broker-dealers wanting
 9   to be dealers in the space.  We actually haven't seen that.
10             What we have seen in our marketplace is the
11   transactional providers, the highly automated firms, those
12   are our largest customers now, and they are really good at
13   providing efficient transactions, but we are losing out on
14   the pricing specialists.
15             We are losing out on the regional investment banks,
16   the retail banks, that provide the price setting
17   functionality, and we need to bring those back.
18             We are big supporters of bringing in increments,
19   but we also think if price increments come in, market makers
20   should be displaying larger sizes.  
21             That is because we should go back to the investor
22   experience.  The investor should start seeing more liquidity
23   offered by our markets at levels at which they want to
24   transact.
25             We run an e-Bay of markets, and right now, the
0031
 1   shopkeeper shelves are too thin for investors.  We need to do
 2   some experiments to try to fix that.
 3             MR. CONROY:  Good morning, everyone.  I'm Brian
 4   Conroy.  I'm President of Fidelity Capital Markets. 
 5   Fidelity, as many of you know, is one of the largest
 6   providers of financial services in the country, with just
 7   under $4 trillion in assets under administration and
 8   management.
 9             We are a firm that is focused on the small
10   investor.  We service tens of millions of retail investors
11   and shareholders in our mutual funds.
12             We greatly appreciate the opportunity to work with
13   the Commission and the staff today, coming from the
14   perspective of the firm, which does not have an economic
15   interest in market structure as we do not provide research
16   and we do not provide investment banking services or collect
17   underwriting fees.
18             Thank you.
19   PANEL 1:  EVALUATING CONCERNS RELATING TO TICK SIZE FOR
20             SMALL AND MIDDLE CAPITALIZATION COMPANIES
21             MR. NALLENGARA:  Great.  I think we are ready to
22   start.  We have a list of questions here that we were hoping
23   to get through.  I am hoping that once we get started, the
24   discussion can flow naturally and we may not necessarily need
25   to turn to each of the questions.  We are hoping we can cover
0032
 1   the topics in the questions by the end of our session.
 2             Taking the questions out of order, to throw
 3   everyone off, I thought it may be helpful to get a sense of
 4   the IPO market today, what impact if any the JOBS Act has had
 5   on the IPO market.
 6             David, you have written extensively on this.  Jeff,
 7   you have a lot of sort of real hand experience.  I am
 8   wondering if you could give us a sense of where you think the
 9   IPO market is now, what impact the JOBS Act has had or could
10   have on that.  Let's start there.
11             MR. SOLOMON:  To start with, certainly doing
12   confidential filing's has helped.  You can point to one area
13   where issuers are more willing to go down the path of being
14   able to file confidentially.  
15             That was a big concern for a lot of people,
16   especially in the tech and health care areas where there was
17   some concern about disclosing company strategies any time
18   before you absolutely need to do that.
19             We have seen a number of companies tap the market
20   using the confidential filing mechanism.
21             It's hard to say that there are more companies that
22   are doing it than would have otherwise done it.  I think if
23   that's the measure for whether or not the JOBS Act is
24   happening, it's still too early to know.
25             The IPO market is still very challenged.  I still
0033
 1   see and speak to a number of people in the venture community
 2   who weigh every day why they would think about trying to
 3   generate liquidity for their investment positions when they
 4   could just go on and sell those companies.
 5             If you think about the strategy that has been
 6   inbred into the source for these new companies, which would
 7   be the venture capital community largely, I think it's still
 8   very much at a minimum, run a dual track to see if I can get
 9   these companies sold.
10             In a worse case scenario, I'm not even really
11   thinking about the public markets because I'm actually from
12   the outset not preparing my companies to go public.  I'm just
13   preparing them to go to a certain size and then sell to a
14   larger aggregator.
15             It will take some time for that to work through.  I
16   don't think the JOBS Act in and of itself will actually
17   change market liquidity or will change a number of issuers,
18   but it certainly is a step in the right direction, and I was
19   a huge proponent of that.
20             MR. WEILD:  I would just add a couple of things. 
21   First of all, 90 percent of the companies qualify as an
22   emerging growth company under Title I of the Act, and 90
23   percent of those seem to be filing as confidential filer's
24   from what we have seen.  To echo Jeff's comments, that's
25   really getting used.
0034
 1             I think the other provision that we find some
 2   excitement with when talking to people on the Street and down
 3   in Silicon Valley is the testing the waters provisions,
 4   specifically for de-risking highly technologically
 5   complicated companies.
 6             In the biotech sector, I've heard some folks that
 7   run equity capital markets, some of the firms like Jeff's,
 8   saying the process at some point going forward won't take a
 9   biotech company public without first investing a lot of time
10   meeting with institutional investors and understanding
11   whether or not the market is there.
12             I think that will cut failure rates over the long
13   run.  It will take a while for that to take hold.  I think
14   that was very beneficial.
15             The rest of it I think is going to take longer to
16   kind of play out.  I think it was Josh that mentioned you
17   need the highways to be attractive because ultimately, once
18   companies go public, they have to sustain issue price.  
19             If they break issue price because of the lack of
20   after market support, economics in the IPO window shut very
21   quickly.  Predictability is not there.  The risk of failure
22   is high.  People start to avoid the IPO market right at the
23   get-go.
24             MR. NALLENGARA:  Much of the current discussion
25   that we are having today -- this is an issue that we have
0035
 1   been talking about for a long time -- much of the issue has
 2   been focused on the IPO market and searching for causes for
 3   the huge decline in the number of IPOs.
 4             Maybe I could ask someone to talk about the impact
 5   of decimalization on the IPO market but then contrast it or
 6   correlate it to other sort of regulatory events or other
 7   events over the last 15 years.
 8             One of the questions that we have had, one of the
 9   questions the Advisory Committee has had, is sort of
10   isolating what impact decimalization or what impact other
11   events have had.
12             MR. KUPOR:  I agree with the general proposition
13   which is there are a number of things that have obviously
14   created this issue, but they all kind of fall, at least from
15   my perspective, under the general rubric of how do we create
16   enough liquidity and enough kind of financial incentive to
17   actually create markets in small cap stocks.
18             Decimalization certainly is a component of that.  I
19   think Jeff mentioned it as did David, global research
20   settlements, very significant in that respect.
21             You really had two kind of very fundamental
22   components that provided most of the revenue and most of the
23   profit making opportunities to support research at the end of
24   the day.
25             Decimalization certainly created profits on the
0036
 1   trading side of things, often used to cross subsidize
 2   research in most of these institutions, and at the same time,
 3   pre-global research settlement, you had a significant portion
 4   of investment banking revenues obviously going also to create
 5   a profit pool that could subsidize research.
 6             I think you have certainly a combination of those
 7   things along with as David mentioned, ATS and others.  
 8             When you kind of bubble all these things together,
 9   at the end of the day what you have done is you have taken a
10   lot of the financial lubricant out of the system for small
11   cap stocks, and therefore, things like research, which
12   obviously helps create and enable trading liquidity, things
13   like profitability at the trading level which incent market
14   makers to actually take risk positions in these companies,
15   the sum total of those things certainly has attributed to
16   drying up a lot of the trading activity and interest in small
17   cap stocks.
18             MR. WOMACK:  I think this issue has focused on
19   decimalization as the problem.  I think we are missing
20   something in terms of the very big picture - the economics of
21   capitalization has changed in the last few years in a pretty
22   big way.
23             I brought a graph which I guess I can't show, but
24   essentially it shows as IPOs have collapsed over this last
25   decade, private equity has taken off.  
0037
 1             If you're looking for something that's more causal
 2   than decimalization, I think it's really more about this
 3   issue of maybe the perfect market that we all wish and hope
 4   was still there, the IPO market is vibrant, maybe it just
 5   isn't the driving force any more.
 6             I have a suspicion that we are trying to make
 7   something happen that isn't there.
 8             With respect to decimalization per se, if I look at
 9   the five or six years before decimalization finally kicked
10   in, the IPO markets were going down precipitously during that
11   period of time.
12             One can't argue that decimalization actually caused
13   that.  It didn't even kick in until four or five years later.
14             I have a healthy suspicion up front before we even
15   start thinking about this that many other things are going on
16   and decimalization is possibly a small part of the fix but
17   not a very big one.
18             MR. GREEN:  I want to go back to a comment that
19   Jeff made earlier, the M&A activity is a source of liquidity. 
20   However, this is all a very careful balance.  We must
21   understand that many of these initiatives that were taken
22   over time, whether they be regulatory, SOX, or otherwise, or
23   were exogenous to that, 2000 .com bust, just market forces
24   themselves, 2008 financial crisis, had an exaggerated effect
25   on the most fragile of all public markets, and that is the
0038
 1   IPO market and the like.
 2             As a result of that, the absence of a viable IPO
 3   avenue has had a significant impact in my view on the
 4   robustness of the mergers and acquisitions market.
 5             It's a check and balance to keep those prices
 6   appropriate that ultimately provide the return to the
 7   investors that make all that work.
 8             My own feeling is that decimalization has had an
 9   impact.  It's one of those factors that is actually quite
10   important to retaining robust public markets, so that becomes
11   an essential part of the ecosystem as this very fragile
12   ecosystem has existed in the past.
13             MR. NALLENGARA:  Professor Womack raised a comment
14   on what may be the real impact of decimalization.  Brian?
15             MR. CONROY:  I think he made some great points
16   about the fact that there are a variety of events that have
17   taken place over the last dozen or so years which have
18   contributed to the decline of the IPO market.
19             We at Fidelity do not feel that decimalization is
20   one of the major contributors to that.  We have not seen the
21   evidence that would indicate in any way it has had any effect
22   on the IPO process itself relative to the others.
23             What we find interesting is most of the discussions
24   are around the fact that the increase in tick size would
25   support the economics of firms to create research for these
0039
 1   small and mid-sized companies at a time when these banks,
 2   these firms, have the ability to raise commissions to their
 3   clients.  This is no longer a net market, this is a
 4   commission based market.
 5             The question we have is if the economic lever
 6   exists for firms to raise commissions to subsidize their
 7   research, to create the opportunity for a more robust small
 8   and mid-cap research marketplace, therefore, IPOs, why isn't
 9   that taking place?  
10             Why are we talking about a pilot program to
11   artificially increase the tick size to create a marketplace
12   which helps market makers at a time when many of these firms
13   aren't even electronic market makers.
14             MR. NALLENGARA:  Professor O'Hara?
15             MS. O'HARA:  Let me follow on Brian's comments,
16   which I think really focus on the fact that as we look at
17   what has happened with decimalization, we have to recognize
18   it changed the picture for everyone.
19             It changed it for everyone in many ways in a good
20   direction.  As we look at transactions costs, no matter how
21   you measure them, transactions costs are a lot lower now than
22   they ever were.  That used to be considered a good thing.  We
23   used to think that was a great idea.  
24             Now it's cheaper for people to trade in these
25   markets.  We generally think that induces people to be more
0040
 1   willing to trade.  Now we are hearing the other side of the
 2   coin, maybe traders, per se, are better off, at least for
 3   listed stocks, and I'll defer to Mr. Coulson about his OTC
 4   stocks, but at the same time, we are arguing that we need to
 5   provide more returns to the market makers, not necessarily
 6   because we want to make market makers better off, but because
 7   we want them in turn to subsidize research.
 8             I think that linkage is really tenuous.
 9             MR. SOLOMON:  Let me just explain a little bit
10   about some of the ways the market economics work.  I would
11   love to go into Fidelity and ask them to pay me six cents a
12   share and not have to worry about it.  The likelihood of them
13   actually doing that is zero.
14             I would argue just like Professor O'Hara says,
15   that's not really the way the world is going.  
16             Fidelity should be trying to reduce its transaction
17   costs as much as it possibly can; it should.  The idea of my
18   going into Fidelity and saying hey, why don't you just pay me
19   six cents, I can write more research.  That doesn't work.
20   It's not practical.
21             What I would say is decimalization, again, one part
22   of it, but a key part to driving research coverage in the
23   after market.  That's what people, I think, don't quite
24   understand.
25             If you can't afford to write research in the after
0041
 1   market, then it's really hard to convince a company they
 2   ought to go public because of the fear there might be a lack
 3   of liquidity.
 4             From our standpoint, I'll speak as a small cap
 5   issuer, Cowen and Company is a public company.  Just this
 6   past week we did an acquisition.  I have two research
 7   analysts that cover us, two research analysts from growth
 8   banks.
 9             I can't get the bulge banks to pay any attention to
10   me because I don't pay them any fees.  They can't make any
11   money trading my 350,000 shares a day at a penny increment. 
12   If you total up the sum total and the spread in my stock over
13   the course of the year, if you had 100 percent of the market
14   share in Cowen stock, you make $700,000.  It just doesn't
15   work.
16             The only reason why people cover us is because they
17   think we might pay them an investment banking fee at some
18   point.
19             Now, one of the two firms that cover us, KBW, just
20   got bought by Steifel.  I can't get the KBW analyst on the
21   phone to talk about the acquisition that we just did that
22   transforms our company.  The only person that showed up was
23   the Sandler analyst.
24             I think there is something more to this than just
25   whether or not we are reducing costs.  I'm all for that.  We
0042
 1   do it.  We have done it on our own.  This is about being able
 2   to provide an ecosystem for an after market, that enables
 3   firms like ours and other firms to cover companies like ours
 4   without having to worry about when is the next investment
 5   banking fee coming to come to pay for that.
 6             MR. COULSON:  As we are talking about having
 7   increments, I don't think the debate should be about having
 8   increments that really artificially widen where things trade.
 9             They should really be about organizing where trades
10   are clustering now.  When you go watch an auction at
11   Sotherby's for a Warhol, they are not running the auction at
12   penny increments for the Warhol. eBay has increments.  
13             Marketplaces use increments as an organizational
14   skill, and that creates meaningful price impact for people to
15   be trading.  
16             We have had a push in our market structure which
17   comes from when Congress brought out the national market
18   system.  Congress said there should be the opportunity for
19   investors' orders to interact without the intervention of a
20   dealer.  They said "opportunity."  
21             We have had a regulatory view that there is
22   something icky when a broker-dealer supplies liquidity to
23   their investors.  There is something bad when you are
24   providing a service that your customer wants.
25             The regulatory process has been trying to turn
0043
 1   broker-dealers into agency brokers going to an order book. 
 2   That is a viewpoint.
 3             We have been slowly -- except for the fastest, most
 4   high tech, whatever increments we want to do, we can go to
 5   .00001 and send it to Knight, UBS, that will be fine.  They
 6   will wiggle around in slippery markets at high speed and will
 7   be incredible at it.  
 8             We are blowing away the price setting world.  We
 9   are blowing away the participants that either through their
10   investment banking, through their institutional sales and
11   trading, through their retail offices -- we used to have
12   Merrill Lynch or A.G. Edwards who would trade a community
13   bank where they had a retail office.  
14             Those were great price setting firms.  They would
15   be there with real liquidity.  The transactional firms will
16   be fine.  We need to bring in those price setting firms
17   again.
18             MR. WEILD:  I wanted to circle back to something
19   that Kent said a little bit about, the problem starting
20   before decimalization in 2001.
21             There were a series of things, if you look at the
22   history, the first thing that put a spotlight on this was the
23   Christy and Schultz study which came out in 1994.  Then we
24   had the Manning rule, which basically said don't trade in
25   front of your customer in 1996.  In 1997 was the order
0044
 1   handling rules, but there was no way to reflect the customer
 2   limit orders into the market until you did it electronically
 3   with Reg ATS.
 4             The seismic shift in the market was really in 1998,
 5   Reg ATS.  Everything was leading up to that.  We went from a
 6   quote driven market where traders could manage their risk to
 7   an electronic order driven market, and what we call the
 8   bankable spread or the maximum opportunity on the desk went
 9   from the maximum quarter point quoted spread to the minimum
10   tick size overnight, which was a 32nd for these smaller
11   stocks or 3.125 cents.
12             As a practitioner running strategy for the firm, we
13   went through a massive realignment internally.  You can
14   remember the SOES, the small order execution system.
15             These things were actually gutting the economics
16   and people were moving out of these markets and trying to
17   adjust.
18             I had a conversation with Dan Case, who was an old
19   friend, who told me when they sold Hambrecht & Quist, he was
20   the President of Hambrecht & Quist, and Dan unfortunately
21   passed away in 2002, but he said one of the reasons that they
22   were selling to J.P. Morgan was they thought in a bad banking
23   market the cash equities business was no longer going to be
24   profitable enough to sustain the infrastructure.
25             SPEAKER:  He was right.
0045
 1             MR. WEILD:  He was right.  I talked to Dan Sturgis,
 2   who was the CEO of Josephthal that did small IPOs.  He's an
 3   investor in the second market.  Dan told me that with the
 4   order handling rules coming in, they knew that was the end
 5   right there of the small IPOs.
 6             Kent said private equity is the place to look.  We
 7   have looked at private equity.  The problem is that private
 8   equity are cash flowing businesses.  If you look at the whole
 9   portfolio of other kinds of businesses, the technology
10   businesses, small IPOs have been sort of eviscerated from all
11   sectors.
12             You could see uptake in the private equity cash
13   flowing businesses and you probably have, but then the
14   question you have to ask is why haven't you seen a sustained
15   small IPO market in other non-private equity sectors, and the
16   answer is you haven't.
17             All small IPOs have dropped like a stone over that
18   period really from sort of 1996 to the culmination in 2001.
19             There was no pilot program done when the order
20   handling rules and Reg ATS were implemented back in 1997 and
21   1998.  This was a wholesale change for our stock markets.  
22             It's amazing to me, and I have to ask the question
23   why it didn't occur, short term measurements, which is what
24   most micro-market economists do, are not helpful when you
25   understand that all the people changes that go on are long
0046
 1   term.  
 2             We don't react over night, firing research
 3   analysts, cutting compensation, because it puts us at a
 4   competitive disadvantage.  This stuff goes on over years and
 5   years.
 6             Just like a buddy of mine who is Professor Robert
 7   Schwartz from Baruch said, stock markets are still adjusting
 8   to changes that occurred in the late 1990s even today.
 9             Take away is just don't look at decimalization. 
10   You really have to look at the order handling rules and Reg
11   ATS.  That is where we saw a very quick drop off in the small
12   IPO.  Ask why we didn't do pilot programs back then, and also
13   recognize that it's not just simply private equity companies
14   that are depressed in small IPOs, it's everything across the
15   board.
16             MR. NALLENGARA:  Professor Womack?
17             MR. WOMACK:  Good points, David.  Before we get too
18   far away from the topic that's come up already, Professor
19   O'Hara and Mr. Solomon mentioned the idea of research coming
20   back into the market through some kind of liquidity provision
21   and profitability, I think that is a really big question.
22             I guess I'm skeptical as to whether we would see
23   small investment banks provide more research coverage if a
24   higher tick size created a more liquid and profitable
25   environment for them.
0047
 1             I've talked to a couple of research directors
 2   recently after I knew I was going to be on this panel to ask
 3   them do you think that's what would happen, if the market was
 4   more vibrant, you would spend more money in research on
 5   smaller companies.  They were quite dubious.
 6             MR. SOLOMON:  I'm a CEO, and what I would tell you
 7   if asked the question how many small cap -- how many non-
 8   large cap stocks do you cover at your firm, which is focused
 9   on financing growth companies?  How many, as a percentage of
10   our research universe?  What do you think the percentage of
11   small cap stocks are that we cover today?
12             MR. WOMACK:  Not very many.
13             MR. SOLOMON:  Not very many.  The reason is because
14   I can't get paid for covering small cap stocks.
15             MR. WOMACK:  No amount of extra --
16             MR. SOLOMON:  However --
17             MR. WOMACK:  From five cents or ten cents is going
18   to change that.
19             MR. SOLOMON:  However, if I saw the opportunity to
20   make liquid markets and reasonable spreads -- we are hiring
21   our first small cap trader.  I'm dedicated to figuring out
22   ways to bring back liquidity, and I'm going to put my own
23   capital at risk in the marketplace to create liquidity, not
24   asking for guarantees, I'm going to put my own capital at
25   risk to try to drive liquidity to names we think are
0048
 1   important to our coverage universe.
 2             As a part of that, we will cover those companies. 
 3   One will follow the other.  It will take some time.  As David
 4   said, if we go through a pilot program, don't make it like a
 5   year because I would assure you that yes, Kent, you will be
 6   exactly right, no one will rush out to re-hire a bunch of
 7   analysts to publish on small cap stocks because it's
 8   expensive to do that.
 9             You need to let that ecosystem regenerate.  When
10   the coverage begins to take hold, you will see liquidity come
11   back to the market because there's no place today for
12   individual investors or small cap institutional investors to
13   find the information they need to make intelligent investment
14   decisions.  It just doesn't exist, for the most part.
15             MR. NALLENGARA:  Brian?
16             MR. CONROY:  I guess I would just shift the
17   conversation a little bit and just remind everyone that in my
18   opinion there has been a paradigm shift in research in this
19   industry.
20             Part of that paradigm shift is because of the
21   decrease of publishing analysts on the Street for a variety
22   of economic factors, many buy side firms, including Fidelity,
23   have greatly grown their research in-house staff.  We have
24   over 100 analysts now on the asset management side focused on
25   following companies.
0049
 1             While we are not publishing, Jeff and his firm will
 2   come in, I'm sure, four times a year and meet with the
 3   analysts and the portfolio managers who are going to make the
 4   investment decisions on behalf of their Funds.
 5             That paradigm shift has occurred.  That is one of
 6   the reasons Jeff suggested firms like ours aren't willing to
 7   pay the six cent commission that we were 20 years ago, and
 8   that we are concerned that if there are changes in
 9   decimalization, increasing the tick size, our transaction
10   costs go up to no discernable benefit for either a retail
11   client or end investor.
12             As far as market makers in this space, most of the
13   traders on the asset management side I'm sure would tell you
14   they don't want partners when buying and selling positions
15   for the firm in these small and mid-cap stocks.
16             We don't need small liquidity events in order to
17   build a position. That just creates more obfuscation in the
18   marketplace.  We would rather it be an agency market, and we
19   recognize that the main reasons there is less liquidity in
20   these are the obvious ones, the float is smaller, position
21   concentration is higher.
22             I think it is important to separate decimalization
23   and the economics of research from this because the world has
24   begun to move on.
25             MR. LEWIS:  I'd like to revisit a point that Kent
0050
 1   made earlier in the presentation.  Before I joined the SEC, I
 2   was a Finance Professor at Vanderbilt University, and I
 3   routinely teach classes on corporate finance.
 4             Part of every one of these classes is a discussion
 5   about the equity offering process.  We go through IPO issues. 
 6   We go through seasoned equity issues.
 7             One of the stylized facts that emerges from the
 8   empirical evidence around IPO issues is the cost to an issuer
 9   of actually going public.
10             Professor Womack noted the rise of private equity
11   as a channel for raising equity capital.  I'd like to hear
12   your comments on the change from using public equity as a
13   source of capital to possibly a lower cost source of capital,
14   i.e., going through the private placement market and the
15   exceptions to the Reg D offering process, if you have any
16   thoughts about that.
17             MR. KUPOR:  I have a couple of comments.  We
18   certainly have seen both in the form of secondary markets,
19   private and secondary markets, as well as new entrants and
20   sources of capital.
21             Even in the businesses that we are in, which is
22   principally financing IT companies, we even see companies
23   like Fidelity and T. Rowe Price and others actually
24   participating in late stage private financing's now, which is
25   something at least we hadn't seen in quite a while.
0051
 1             We have also seen other alternative sources of
 2   capital like secondary markets.
 3             I think one interesting policy question that it
 4   does raise for the SEC consistent with protecting investors
 5   and ensuring fair markets is are you at risk of potentially
 6   having more of a two tier market system over time, which is
 7   you have private equity investors like ourselves, you have
 8   kind of secondary equity investors who are accredited and
 9   therefore obviously can participate in those markets, and yet
10   the public markets where you have individual investors being
11   able to actually trade those stocks continue to be much less
12   accessible to higher growth companies.
13             One of the things certainly that we see and you see
14   this in a lot of the numbers is an elongation of the time to
15   going public for these companies which in general also means
16   that the growth trajectory and the kind of appreciation
17   opportunities in the public markets are probably less so than
18   they were when we had a more robust small cap market.
19             I agree with the general statement which is I think
20   there are secondary sources of capital coming into this
21   market.  I think there is a bigger policy question
22   potentially for the SEC and other organizations as to what
23   the implications of that are if we don't enable some of these
24   companies to go public at a time which is early in their life
25   cycle where there is more growth trajectory and therefore
0052
 1   hopefully appreciation for public market investors.
 2             MR. GREEN:  I would add a couple of things to that. 
 3   One is what I'll call the growth equity sector, not the pure
 4   private equity sector, but the growth equity sector is a
 5   reasonably small amount of the total capital markets in terms
 6   of the availability of capital.
 7             I don't see that changing exponentially over the
 8   near term because the opportunities still needs to
 9   understand, the companies themselves.  It requires a
10   significant amount of due diligence, relationship building
11   and all the rest of it.  It's much more of an individualized
12   dance.
13             The public markets by contrast, notwithstanding the
14   cost to go public, provide a far lower cost of capital than
15   growth equity will in terms of raising that money.
16             Not only that, but enables the fundamental issue of
17   liquidity, and that promise of liquidity is actually one of
18   the major motivator's that a private equity or growth equity
19   investment cannot assert.
20             We actually have now a subcommittee in the National
21   Venture Capital  Association on growth equity, to help
22   stimulate that and encourage it.  We do think it's a sector
23   that needs more encouragement.
24             I don't think it will be able to become -- just in
25   terms of dollars alone in the foreseeable future -- anything
0053
 1   close to what the reasonable potential is of the public
 2   markets to provide capital.
 3             MR. NALLENGARA:  Gregg?
 4             MR. BERMAN:  Can I go back to a couple of comments
 5   about the implication of changing tick sizes to something
 6   that is larger?  
 7             One of the common refrains that we hear is I can't
 8   buy stock, it's illiquid because the spread is too high
 9   already.  For many small and mid-cap stocks, they don't trade
10   the penny.
11             The actual physical quoted spread might be seven
12   cents, eight cents, 15 cents, 20 cents if it's an extremely
13   liquid stock.
14             What aspect of changing the tick size would either
15   stabilize that size or can you talk a little bit about why
16   the existing wide spreads for illiquid stocks don't seem to
17   provide the motivation that's needed to gain the profits and
18   produce that research.
19             MR. COULSON:  I think the real issue is if you
20   don't organize it into increments, there's not as big an
21   incentive to show any size because someone is just pennies
22   ahead of you for 100 shares, and marketplaces, there is a
23   slight view that marketplace prices are set by people who
24   make offers.  They make a bid.  They make an offer.  That's
25   not really true.
0054
 1             Market prices are set by people who take offers and
 2   hit bids.  There are lots of people who have houses for sale
 3   and get an offer that's ludicrous.
 4             What you want to do is incentivize the display of
 5   liquidity and you want to make for that security the two --
 6   for someone to move ahead, that they have to move to the next
 7   price point.  It shouldn't be an artificial price point. 
 8   That shouldn't be the idea of increments.  It should just be
 9   the next price point.
10             There are community banks that are actively traded
11   on our market, trade 30 times a day.  There is almost 30,000
12   seconds in a trading deck.
13             There is a lot of time in between those price
14   points.  It's pretty easy to go through and say what are the
15   price points of these classes of securities.  
16             There is a great paper by AMF, which is probably
17   the smartest thing I've seen on this and explains it better
18   than I ever could.  It comes from the French, so it's a
19   little complicated.  We probably use a little brutal
20   simplicity in America.
21             The idea is we should have price points based on
22   price and trading velocity and number of trades in a deck. 
23   That organizes it.  It brings together the structure, and
24   that way, the next price is something that has meaning.  It
25   has economic difference.
0055
 1             For something that trades 10,000 times a day, that
 2   next price is very tiny, that next increment, but we stopped
 3   organizing it for smaller companies.
 4             MS. O'HARA:  I want to pick up on comments Gregg
 5   led us to and Cromwell has focused on as well.  It is
 6   certainly true there are stocks that have wide spreads, but
 7   we also have to be really careful, there are lots of small
 8   stocks now that trade with very, very small spreads.
 9             For example, my colleague, Gideon Saar at Cornell,
10   we were chatting, and he gave me some numbers.  I think they
11   are interesting.  These are from NASDAQ.  We are talking
12   about stocks that are now listed.
13             If you look at stocks that trade less than $250,000
14   a day in value, which are fairly illiquid traded, 469 of
15   those, but 211 of them have spreads that are less than a
16   nickel.
17             If you now start putting nickel spreads on these
18   stocks, you are essentially by definition just raising the
19   transactions costs for everybody else relative to where they
20   are now.
21             If you say let's restrict it to stocks that trade
22   $100,000 a day, which really is not a whole lot, there are
23   257 of them and 108 of them have spreads less than a nickel.
24             Even going down and saying what if they have less
25   than 100 in market cap, you can have 95 of them that have
0056
 1   spreads less than a nickel.
 2             I think what we are hearing is a Tale of Two Cities
 3   almost here.  It is true there are stocks that have high
 4   spreads, but I think one of the things that is difficult and
 5   certainly things that we as researchers find is it's not as
 6   straightforward as it looks.
 7             You can't just say look at all low priced stocks,
 8   they have terrible spreads because a lot of them don't, even
 9   stocks that aren't trading as much as you might expect, some
10   of them have very low spreads.
11             As we begin to change these tick sizes, I think we
12   have to recognize the complexity of what we are doing here.
13             You are not just painting the hole in the wall. 
14   You are painting the whole wall.  There are all kinds of
15   problems that I think arise when we do that. 
16             MR. COULSON:  There are 600 community banks on our
17   platform which I think is a great example.  They have
18   averaged -- the last I looked, the spreads were about 20
19   cents.  
20             If your kind of average spread is 20 cents, you are
21   not really incentivized to show much size if someone can pop
22   ahead of you for 100 shares and make the spread 19 cents.
23             It should be a stock by stock basis.  I think the
24   greatest thing is -- I'd love to have a pilot program run
25   across our market, and let companies opt out if they want,
0057
 1   and that is I think we would see demand, and there would be a
 2   great test on markets.
 3             Markets should set what is the increment of quotes
 4   and orders.  Trades should take place any place.  Actually,
 5   for institutions, they should never trade on the bid and
 6   offer.  They should try to trade as close to the middle as
 7   possible so they are not giving away information which way
 8   the blocks are going.
 9             We shouldn't be forcing institutions to have to
10   trade on bids and offers, but we should be setting those
11   increments and having the markets stack up.
12             We have had a trend.  It used to be investors
13   traded with marketable orders, and because there was on the
14   offer that price.  Now as markets have gotten slippery, you
15   have to have a limit order.  The investor comes in and the
16   idea that a retail investor has market impact is slightly
17   ludicrous now.
18             MR. SOLOMON:  If you are a retail investor, if you
19   ever asked your retail firm to give you a volume weighted
20   average price execution over the course of the day, they will
21   look at you funny.
22             I did that last week actually.  I did something I
23   haven't done in a while, I bought a stock.  It's hard for me
24   to buy a stock because of a lot of the compliance rules, but
25   I did it just to see what would happen.
0058
 1             I put in an offer to buy a stock that was trading
 2   at $13.50 by $13.51, and before the open, I put in to buy
 3   4,000 shares of a stock that I knew had averaged around
 4   200,000 to 250,000 shares a day.
 5             Not inconsequential as a percentage of the daily
 6   average trading volume, and I asked them to give me a volume
 7   weighted average price between now and the end of the day and
 8   then allocate it to three different accounts, which is
 9   basically trader speak for how I used to trade my
10   portfolio's.  
11             I want to buy over a period of time and I want to
12   allocate it down.
13             By the time the market opened, the stock was
14   trading at $13.66/$13.67.  I called them back and I said do
15   me a favor, I was looking for the volume weighted average
16   price over the course of the day, I hope they're not walking
17   up the bid because they see a 4,000 share order on the desk.
18             They assured me that wasn't the case and they
19   called me back at 9:28 and said to me oh, we can't do volume
20   weighted average price.  We need a limit order.  I'm like
21   really.  A limit order, what's that?  That's code for I want
22   to rip your face off.
23             Fine.  I gave them a limit order at $13.68.  Don't
24   you know, the first print, $13.68.  Three minutes later,
25   $13.47.  Really.
0059
 1             I called them on the phone.  That $13.68 print
 2   better not be my print, because if it is, I'll start an
 3   arbitration just for laughs.  It doesn't matter to me, it's
 4   4,000 shares, it's 20 cents.  It's not going to make or break
 5   me.  
 6             When we are talking about issues, about protecting
 7   retail investors, and there is a lot of discussion in here
 8   around how decimalization impacts costs, and yes, it's true,
 9   but there is a lot more things that go on in the market that
10   disadvantage retail investors than penny increments.
11             There is a lot that the SEC could and should be
12   doing to actually open up order types for retail so they can
13   trade like everybody else does.
14             I think that is a key thought, if there is a lot of
15   push back around running a pilot program, let's think about
16   other things.  I think Professor O'Hara is right, we should
17   be thinking about other things like addressing maker/taker
18   rules, that just level the playing field for the average
19   investor so that you can begin to have a little bit more
20   price discovery, fundamental price discovery, rather than
21   casino type transactions.
22             Those are things I would say are far more
23   fundamental to investor protection than whether or not we are
24   trading in penny increments.
25             MR. CONROY:  I would agree.  I think it is a
0060
 1   marketplace that has become more sophisticated and not all
 2   participants in the marketplace have kept pace.
 3             The typical retail investor is not buying 4,000
 4   shares in a small or mid-cap stock.  They are buying much
 5   smaller share sizes and can take advantage or be advantaged
 6   by the smaller tick increments.
 7             I think the issue is making sure that the investors
 8   who are participating in the marketplace, if they are not as
 9   sophisticated as Jeff is, should become some, and tools being
10   provided to them.
11             MR. WOMACK:  Gregg, I wanted to give you an example
12   about pre-Reg ATS, how we traded stocks and moved them
13   through.
14             Just say a stock was trading 10 by 10.25 with a
15   quarter point and it was a quote driven market.  We can buy a
16   block of 100,000 shares at 10, mark it up with a quarter,
17   sell it in 1,000 share increments at the retail system or
18   call institutional accounts.  
19             We could make as much as 2.5 percent.  We could
20   turn around and print it in the middle up an eighth and make
21   an eighth of a point to Fidelity.
22             As soon as we went into the Reg ATS post-world
23   market, and when people progressively started to be able to
24   step in front of you for a penny, all of a sudden, we
25   couldn't count on an economic opportunity, so we couldn't
0061
 1   commit that capital.
 2             All the capital came off the desk for these small
 3   cap stocks.  
 4             Brian, you made some points earlier.  We do
 5   analyses on the institutional marketplace.  I will tell you
 6   about 80 percent of the commission opportunity rests with
 7   about 100 institutional investors in the United States. 
 8   Fidelity is at the top end of the curve, very close to the
 9   top end of the curve.
10             The difference in terms of assets under management
11   top to bottom, the 50th account is about 1/20th the size of
12   the largest account, $50 billion versus $1 trillion, up at
13   the top end of the curve, Fidelity, BlackRock, T. Rowe Price.
14             If you look historically, going back 15 years ago,
15   you will find that all the largest accounts have cut as a
16   percentage of their assets under management their allocations
17   to small cap, micro-cap and nano-cap stocks.
18             If you look down the long tail of the distribution
19   curve to the 2,000th largest investor, you will find they
20   have increased their allocation to small cap, micro-cap. 
21   That's the self directed market.
22             What has happened is you have gutted the economic
23   incentives for people to distribute securities so the breadth
24   and the quality of the distribution of Wall Street on IPOs
25   has declined.  
0062
 1             The after market distribution has declined because
 2   there is no longer any economic incentives, and Wall Street
 3   has reacted by concentrating its resources on a smaller
 4   number of institutional accounts, it has closed down every
 5   middle market institutional sales desk.  
 6             Every firm had them.  We had one at Prudential
 7   Securities.  Gone.
 8             Retail brokers used to actually cover smaller
 9   institutional investors that were not so into the
10   institutional marketplace and they would be supported with
11   capital from the NASDAQ or the upstairs New York Stock
12   Exchange trading desk.
13             This has been a massive deterioration in
14   distribution fundamentally.  What we see is IPO sizes have
15   gotten larger, but IPO success rates have declined steadily
16   over 15 years.
17             That's very scary to me.  It's gone from about 75
18   percent the way we define them, 70 percent of all IPOs were a
19   success, going back into the early 1990s, to about 30 percent
20   today.
21             As stocks are breaking issue price because the
22   distribution support isn't there, they are more mature so
23   they should be better companies, why do you think we are not
24   getting more IPOs done?  Because investors aren't making
25   money because the deals aren't being supported.
0063
 1             MR. CONROY:  I would say while that was all
 2   occurring, marketplace has shifted from a net market to a
 3   commission based market.  In this market, brokers do have the
 4   ability to charge more commissions per trade.
 5             The economic forces of the buy side are preventing
 6   them from doing so, as Jeff pointed out earlier.
 7             I'm not quite sure why the industry should be
 8   supportive of increasing tick sizes and going back to try to
 9   relive a marketplace that is a spread market which no longer
10   exists.
11             MR. SOLOMON:  I think it is becoming increasingly
12   difficult for the buy side to out perform indices, and I
13   think we hear that every day from clients, like Fidelity,
14   that it is really super hard now to provide incremental alpha
15   above the beta benchmarks.
16             The easiest way actually to out perform benchmarks
17   is to participate in new issue.  
18             I'm not surprised to find retail investors and
19   investors at large choosing to invest in ETFs and lower cost
20   investment vehicles because they are seeing interestingly
21   enough that the money that they are paying to act as managers
22   isn't really worth the return they are getting over and above
23   the benchmark.
24             I would say we are actually on the same side here
25   in that regard.  A healthy IPO market provides long term
0064
 1   active managers with positive returns over the long haul.
 2             I'm not interested in going back to Fidelity and
 3   asking them for six cents.  That's not the point.  The point
 4   here is we have a responsibility to protect the ecosystem
 5   that allows companies to become public, and that means
 6   driving liquidity back to smaller issue where the bulk of
 7   these companies come, and decimalization is one very
 8   important piece to that, and widening tick increments
 9   incentivizes people to actually do fundamental price
10   discovery rather than electronic price discovery.  They are
11   two very different things.
12             What has also gone on here is I think there is the
13   maker/taker dynamic that goes on in the market and that
14   provides for a significant amount of volume that has nothing
15   to do with price discovery.
16             It is just how fast can you get that execution done
17   so you can be the maker and get the rebate, and somebody else
18   is the taker.
19             It is a zero sum game, by the way, between makers
20   and takers, and the people taking the spread out of the
21   middle are the people that are running alternative trading
22   venues.  
23             They are not really interested -- they are
24   interested in order matching at this point.  There is really
25   no price discovery in order matching.  It is just I want to
0065
 1   be the biggest.  I want to have the most flow.  I want to
 2   have the most connectivity because I can make the penny
 3   increment between the maker and the taker.
 4             That kind of behavior, I'm not saying it's the root
 5   of all evil, but it does need to be tailored and dialed back
 6   a little bit when you're talking about small companies where
 7   fundamental price discovery is so important to creating
 8   functioning after markets.
 9             COMMISSIONER GALLAGHER: I have a question for
10   David.  This conversation is fantastic, by the way.  Self
11   sustaining, easy for the moderators here.
12             I think David hit on something that I've been
13   thinking a lot about and talking about publicly, which is the
14   fear at least I have on the Commission, that we are servicing
15   a symptom and not the illness.
16             Decimalization could be simply a symptom, which is
17   the point I was getting at in my opening remarks.
18             I'm fascinated by this discussion.  The order
19   handling rules.  People don't talk about the order handling
20   rules.  I've been talking about -- to Cromwell's point -- the
21   1975 Act amendments forward.  
22             It's about time to question these things.  If not
23   change them, at least severely question them before we start
24   treating more and more symptoms and making things more and
25   more complicated.
0066
 1             We have a pretty complicated carpet we have woven
 2   here on market structure.  
 3             Specifically as to your example about pre-order
 4   handling rules, how a big block would have been treated, do
 5   you have a recommendation in that space?  
 6             Should there be a carve out within the order
 7   handling rules for certain sized companies?  Is there
 8   something else we should be thinking about in that vein?
 9             MR. WEILD:  The genesis of the tick size idea was
10   at a dinner a year and a half ago with a bunch of market
11   professionals, former Stock Exchange executives, and sell
12   side and buy side folks.
13             People just thought it would be easy to implement. 
14   Part of it is if you were given a clean piece of paper and
15   you wanted to redesign the entire market, what would you do.
16             I'm a big believer that you shouldn't get in front
17   of your customer.  The flip side is when you have to reflect
18   every single order into the market and then you start to
19   proliferate the number of price points, the real economic
20   incentives and the ability to manage risk that traders used
21   on trading desks was a function of the reliability of the
22   spreads.
23             To Maureen's point, even if something is at a
24   nickel or 20 cents, the fact is it's illusory.  I can't bank
25   that spread.  I'm not going to commit capital. 
0067
 1             I think if you can create price increments which
 2   are larger -- as Cromwell has pointed out, they sort of
 3   naturally -- remember, the penny tick is not natural.  We
 4   ordained it.  We made a determination. 
 5             If you go back, this goes back to the discussion
 6   around the order handling rules and Reg ATS.  When those were
 7   being implemented, if you go back to the record, the SEC
 8   decided to do it anyway, they had received quite a bit of
 9   feedback from the Street that it was going to harm capital
10   formation.  They believed otherwise.
11             All we are talking about is adequate economic
12   incentives for small cap companies.  You don't need it for
13   large cap which trade, as the academic's would say, these are
14   symmetrical order book marks, with lots of buyers offsetting
15   lots of sellers, naturally visible, on everybody's radar
16   screen, they benefit from the network effect, but small
17   company stocks need to be sold.  They need to be marketed.
18             If there is no economic incentive to do that, a
19   $250 billion market value company is 2,500 times larger than
20   a $100 million market value company.
21             Yet, the basic economic model that we have created
22   is the same for the one that is 2,500 times larger than the
23   little one.  On its face, that just doesn't make any sense,
24   if you're trying to optimize markets and create balance in
25   the system.
0068
 1             I think we have to be very careful of ideologies
 2   that favor low cost investment when in point of fact, as
 3   Warren Buffett has said, you know, it doesn't bother him to
 4   pay more on a transaction because he's a long term investor.
 5             What we are catering to is trading.  As a result,
 6   we have seen more and more coverage of the Street's shift
 7   from investors to the traders, because there is where the
 8   money is, which is exactly the people from a balance
 9   standpoint, the investors, with some exceptions, that issuers
10   don't need to reach.
11             One of the unintended consequences is if you look
12   at the client access business, non-deal road shows, Wall
13   Street is wasting the time of issuers, running them around in
14   front of institutional investors who are not natural buyers.
15             Many times I see this with our clients.  They are
16   smaller cap, they are in front of somebody who can't possibly
17   own a $500 million equity market value name.  
18             Why?  Because they're an expert in a vertical,
19   providing information about their competitors in that
20   industry to an institutional investor that is the client of
21   Wall Street.
22             There are a lot of things that have effectively
23   shifted all sorts of work onto the backs of issuers.  Issuers
24   are woefully unprepared, and as a consequence, they are
25   floundering as public company executives because they don't
0069
 1   come to it with the knowledge base that the panelists do
 2   here.  They are ill equipped.
 3             I think we just need to get adequate economic
 4   incentives, not to gouge people.  Time and price priority,
 5   very important.  As an investor, in Jeff's example, if he's
 6   first at a price point, if he's paying a nickel or dime,
 7   whatever the case might be, he should be able to get up, if
 8   he's first in line, he should get the execution, and that
 9   instills enormous amounts of confidence to retail investors
10   because it creates a sense of fairness.
11             Right now, this cat and mouse stuff that goes on,
12   you know, where people are putting in trades for retail
13   investors, undermines investor confidence, including my
14   friend, Bob Aber, who used to be general counsel of NASDAQ,
15   who is an active trader.
16             He always complains about the fact that as an
17   active investor, when he puts his little orders in, somebody
18   always automatically pulls him out, and then the order in
19   front of him disappears.  Bang. Bang. Bang.
20             I think we can do a lot by just bringing it back to
21   the center a little bit and making sure there is an effective
22   seat at the table for the issuers that need to have a
23   reasonable experience as a public company and the large
24   investors, but also to understand that distribution, there is
25   an enormous amount of small cap distribution in this country
0070
 1   that is not being accessed because nobody can afford to cover
 2   it any longer.
 3             MR. COULSON:  In some ways, we're not a very big
 4   IPO market.  Actually, when David was Vice Chairman of
 5   NASDAQ, they did become a public trading OTC first, and then
 6   they did a secondary on NASDAQ.
 7             It's really about creating the quality of market
 8   long term.  We have two separate subjects.  If we have a
 9   quality of market, and our markets, we are a significant
10   source of Exchange upgrades, about 40 last year when to
11   NASDAQ and NYSE, and bringing companies in, and getting that
12   conveyor belt.
13             The JOBS Act, what has the JOBS Act done?  I ask
14   when is the JOBS Act going to be done because in the smaller
15   company financing space, it's all Reg D.  We're not done on
16   that.  We have not even started on Reg A, 144(a).
17             Those are the things that are going to go create
18   equity that can then go be tradeable, and equity is worth
19   more if it's tradeable.  If it's only a tiny amount and the
20   investor experience is not good -- investors want to be able
21   to trade.
22             Just to address, I do have a disagreement about the
23   idea that the person who makes the offer should be first in
24   line.  
25             When I go buy an iPhone and I show up at the store
0071
 1   with an ad of an iPhone offering $25 cheaper, I don't go to
 2   the store guy and he says okay, I'll match that price, I
 3   don't say oh, no, I want the phone in the ad.  Can you go to
 4   California and get that phone and that guy who made the
 5   offer.
 6             This is how competition -- the taker, the consumer
 7   should be the one with the choice, they should get the best
 8   price but they should choose based on the quality of service,
 9   based on where they go, and that incentivizes more aggressive
10   off bids and offers.  
11             That should stay in our markets.  We really have a
12   dynamic market eco-structure where it's not one size fits
13   all.  We have different types of brokers that fit large
14   institutions, liquid, and that is incredible.  
15             Sadly, they don't cross every trade Fidelity has. 
16   In fact, the majority of institutional trading flow is now
17   being broken up into tiny retail sized bits and filled by a
18   liquidity provider.  That's a little bit of a dark secret
19   that liquidity providers are still very important in this
20   agency market.
21             It is not to create a disadvantage for dealers but
22   for us to go back and have fair competition between brokers
23   and dealers.  The Doctrine of the Church has been read that
24   we should prefer agency brokerage.  Being a dealer is icky. 
25   Supplying liquidity puts you in a questionable position, even
0072
 1   if you are supplying liquidity at the best price.
 2             I think we should bring it back and just have it be
 3   fair and organize the market so it works for agency brokers,
 4   it works for dealers.
 5             Our marketplace, we have agency brokers, we have
 6   NYSE's ArcaEdge ECN, who does about eight percent of the
 7   share volume, and we have market makers.  They all interact
 8   and compete and investors get to choose.
 9             That's what we want.  We want equal competition,
10   not an one sided push, you should become an agency broker,
11   and then the institutions will pound you on per share
12   trading.
13             COMMISSIONER GALLAGHER:  Thanks, Cromwell.  I just
14   want to pick up before we end this panel here on a last point
15   David made on research, that we aren't going to have
16   liquidity.
17             I want to just challenge that point because this
18   came up in our Small Business Committee meeting last week,
19   why are they recommending to us to do this pilot on tick
20   sizes.
21             There was actually some debate.  Was it research or
22   was it simply just liquidity, was it the ability to make the
23   spread.
24             I'd like to just throw that back.  I don't know,
25   Kent, if you have a thought on that.
0073
 1             MR. WOMACK:  I was actually thinking about that
 2   just this second.  It strikes me that these things are very
 3   separable, and what we have been talking about for the last
 4   10 or 15 minutes here is improving the liquidity environment.
 5             I just don't -- I'm much more skeptical that we can
 6   put a gun to anybody's head and say produce more research.  I
 7   don't really see what the mechanism is for driving a larger
 8   amount of research, will in fact the small brokerage firms
 9   with that liquidity provision throw off some of it for
10   research.  
11             How are we going to know?  I don't know how we are
12   going to test that very easily in any reasonable length of
13   time.
14             MR. SOLOMON:  I would just say from my standpoint
15   as somebody who can make that decision every day, we are
16   taking companies, small companies, on the road and doing non-
17   deal road shows all the time.  
18             We do take them in to see large buyers who we know
19   can necessarily accumulate a position because there is a lot
20   of transmission of knowledge about what is happening in that
21   particular industry, but we are also taking them out there
22   because we do need to cultivate a very personal relationship
23   between those management teams and the ultimate end buyers,
24   and there are a number of them out there.
25             What I would say is a quintessential part of being
0074
 1   able to get in the door to have those meetings is published
 2   research.
 3             We have turned our business model around and said
 4   corporate access is not just an administrative back water
 5   where you are filling meeting calendars between companies and
 6   potential portfolio managers.  It is actually a driver of
 7   value and connectivity between the end buyer, the investor,
 8   and the management team.  
 9             You want that conversation to happen as many times
10   as you possibly can so the investor is better informed and
11   the management team actually appreciates what the owners in
12   the company want him or her to do in that position.
13             Writing research gets you the ability to open the
14   door at the buy side.  If you don't write research, it's just
15   hard to get in the door because they are like why are we
16   talking about this.  You don't write on it and I can't buy
17   it, why are we talking about it.
18             I start with my research report.  Here's our view
19   on this company.  We are bringing management into Boston and
20   we would like them to see you and here's the reasons why, in
21   three bullet points on the front page.  Can we see you for 15
22   minutes.
23             That's why research is important and it drives the
24   back end of the business which ultimately leads to capital
25   raising and ultimately leads to IPOs and job creation. 
0075
 1   That's really the connectivity here.
 2             COMMISSIONER GALLAGHER:  Brian, on that point, on
 3   the buy side, the utility of the research, because that is an
 4   interesting point.
 5             MR. CONROY:  As I stated earlier, we, too, are
 6   skeptical that somehow decimalization is linked to the
 7   economics of research, especially in a time when we are no
 8   longer a net market, we are a commission based market. 
 9   Brokers do now have the flexibility to raise commissions.
10             We think the real issue for looking at IPOs is 20
11   years ago the average underwriting spread for a small or mid-
12   cap company was 500 to 700 basis points.  Today, it is
13   exactly the same.
14             If you look at how market structure has improved
15   liquidity, quote size, reduced volatility across the board,
16   what stands out unique in our market is the fact that the
17   underwriting spreads and the lack of transparency in equity
18   issuance has remained where it has been over the last 20
19   years.
20             I think one of the reasons the buy side is
21   skeptical about raising spreads to subsidize research would
22   be where is that 500 to 700 basis points going now.
23             MR. KUPOR:  I agree with Jeff, and David, you may
24   agree or not, but I think even if you didn't get research, if
25   you got better trading liquidity in these stocks, to David's
0076
 1   point, making sure it didn't just take a penny increment to
 2   step ahead of price and you incented people to actually have
 3   larger trading volumes, I think that would be a major push
 4   forward here in terms of helping the trading liquidity of
 5   these stocks and then enabling them to go public but also to
 6   have a robust trading environment that allows them to do
 7   things like secondary offering's to raise capital in the
 8   market.
 9             I take Jeffrey at his word that providing financial
10   incentive would translate into research.  I think even in the
11   absence of that, if you solved the liquidity problem
12   fundamentally through the tick sizes, that still has a
13   tremendous impact on the attractiveness of the capital
14   markets for small cap issuers.
15             MR. GREEN:  Yes, stated just slightly differently,
16   liquidity is the major point.  Research, in my view, is a
17   tool to increase liquidity.  Therefore, in stimulating that
18   and all the rest of it is -- I don't want to say it's a side
19   show, but it is not the main event.  The main event is
20   liquidity itself.
21             I just can't help but say since I have the mike for
22   maybe the last time here, just a couple of thoughts as you
23   design a pilot program moving forward, there is a tendency in
24   the law and regulation to design these like engineering is
25   designed, and this is relating back to where I live in
0077
 1   Silicon Valley.
 2             Therefore, very discernable, a pragmatic type of
 3   approach and the rest.
 4             I would consider these pilot programs to be more
 5   like biology.  We don't really know what the results are
 6   going to be.  We are going to try certain things.  We are
 7   going to try to get end points.
 8             What biologists do is try to get end points from
 9   their experiments and figure out what that is.  For me, it is
10   liquidity.  That is the bottom line.
11             What does liquidity really mean?  I divide it into
12   three different areas. One is what does it do to pricing,
13   what does it do to volume, and then finally, is it
14   encouraging investors as opposed to trading.  Does it mean
15   trading is unimportant, in fact, it's absolutely critical. 
16   Is it emphasizing more investment in small cap stocks.
17             MR. WOMACK:  At the same time, I think factoring in
18   the incremental costs to the investor in moving this tick
19   size, we are completely in a cost/benefit tradeoff situation
20   here.
21             MS. O'HARA:  I still think we need focus, as you
22   pointed out, if the focus is on liquidity as opposed to
23   research, then I think we need to use the data that we have
24   to help us here.
25             The data that we have actually suggests that the
0078
 1   transactions costs, the trading costs of trading small stocks
 2   has actually been falling dramatically.  
 3             ITT Research, Jackman King, put out a very nice
 4   piece showing, for example, that trading costs for small
 5   stocks decreased at a faster rate than it did for large cap
 6   stocks from 2008 to 2012.  
 7             I really like the way you focused our attention,
 8   that there are two issues here.  There is a liquidity issue
 9   and there is a research issue.  
10             I think one of the interesting challenges that the
11   SEC will have to think about is what is going to be the point
12   of your pilot.  I think there can be different goals and
13   different directions that this can go.
14             COMMISSIONER GALLAGHER:  Just to be absolutely
15   clear, despite the nodding of heads over here, I don't think
16   we should assume a pilot.  I think the Commission has a lot
17   of hard decisions to make here as to whether this makes sense
18   to proceed, to the discussion we had with David, are there
19   other things we should be thinking about.
20             MR. WEILD:  Personally, I'd just do it.  That's me.
21             (Laughter.)
22             MR. WEILD:  They did it back with the order
23   handling rules and Reg ATS.  I think it's about visibility
24   and creating and sustaining visibility and driving down cost
25   of capital for small cap stocks at the end of the day.  They
0079
 1   have to have competitive costs of capital.
 2             I think research -- there is a great paper that was
 3   just put out by Tim Keating at Keating Capital that just
 4   charts the lack of equity research coverage.  That's helpful
 5   to the process, but salesmen, sales traders, they make phone
 6   calls.  They create the visibility and the liquidity.  The
 7   capital committer's facilitate that as much as the research
 8   analyst.
 9             A lot of that research is even worse.  Most of that
10   small cap research isn't being actively marketed anyway
11   because there is no sales incentive to do it.  It's a place
12   holder to keep the client happy because they may do an
13   investment banking transaction at a later date.  
14             It's very misleading.  I would say you just have to
15   get people focusing on these stocks.  You have to get it to
16   the point where Fidelity will reverse the trend they have had
17   for the last 15 years, which is cutting their allocations as
18   a percentage of their portfolio to small, micro and nano-cap
19   stocks, and actually starting to raise it because they can
20   get a bid.  There is liquidity for them to put size to work.
21             If we start doing that, we will get America back
22   into business.  We will bring back $25 million IPOs again. 
23   We will start to create jobs and reverse this 15 year
24   decline.
25             MR. COULSON:  The electronic market makers, the
0080
 1   highly automated ones, I have not heard concerns of
 2   profitability in trading the small cap stocks on NASDAQ
 3   because they are great at wiggling around.
 4             Every study is a series of statistics, what's the
 5   spread, but what is the effective spread, what is the wiggle. 
 6   Where is the satisfaction.  Are we designing -- in our debate
 7   with FINRA over new tier sizes in our marketplace, the real
 8   debate was who was more important, the 500 share investor of
 9   a $6 stock or the 500 or the 100 share investor.
10             I think we need to put it in a little bit of the
11   middle, that there is enough liquidity that there is not
12   these 100 share spreads, that it starts to stack up.
13             On the other point, we have been building an
14   ecosystem at the top level of our marketplace with OTC, you
15   actually need to be sponsored.  We have used attorneys which
16   do a passive quality check, but we also have investment
17   banks.
18             What we have seen is research is one side, but the
19   real value is having an institutional firm take you on the
20   road, take you to conferences, be there as a call to trade
21   the stock when a block comes up.  It's that sponsorship.
22             In Europe, we have the nomad corporate broker
23   bringing in that institutional connection to the markets for
24   our company, because we are creating these orphans. 
25             If you see a lot of research, it can be good and
0081
 1   bad, but you really need corporate coverage, you need someone
 2   on the institutional desk who knows the company, knows the
 3   story.
 4             In our market space and in the bottom half of
 5   NASDAQ, people don't do secondaries and capital raises.  They
 6   are doing private placements.  They are in the pipes market. 
 7   They are going to a few sophisticated investors.
 8             What you really need if you are in the capital
 9   raising mode as a public company is to be out there talking
10   to these investors before, because when you do a capital
11   raise as a pipe, you get such a better capital raise if you
12   have knowledgeable investors.  If they are fundamental
13   investors rather than someone who is trying to just arbitrage
14   it into the market.
15             That part of connecting, we have a problem if it
16   just becomes all these structural arbitragers for the small
17   company capital raises doing a pipe, rather than we have
18   institutions that are fundamental investors, no companies
19   that corporate brokers have taken them out to.  That is part
20   of the connection of getting these firms back trading these
21   stocks.
22             The high speed firms, they are fine.  Make it
23   whatever you want.  Set the rules.  They have great
24   technology bases.  The big five/ten of those will just move
25   to trade whatever type of security.
0082
 1             We want to bring back the ecosystem to the price
 2   setting, the corporate financing, and it's about quality of
 3   market experience.  If we have quality markets, we will get
 4   more publicly traded companies.
 5             MR. WEILD:  Chris Jacobs, who was the co-Chairman
 6   along with Steven Graham of the SEC Advisory Committee on
 7   Small and Emerging Companies that made the recommendation on 
 8   Friday to increase tick sizes, she took her stock chart back
 9   when the order handling rules and Reg ATS were implemented
10   and just overlaid it, and found that even though she was
11   growing at 30 percent, compound annual growth rates dropped
12   off the cliff.
13             It might be helpful for people to take a look at --
14   that's just one data point -- what happened around that
15   particular period.  That was a shift in stock markets.
16             All the focus of the academic papers that I have
17   read have been around decimalization.
18             MR. BERMAN:  Could I ask just a quick question?  We
19   have heard a lot of numbers in terms of what is a small cap. 
20   I know that from a Fund standpoint, from an investor, if I
21   invest in a small cap fund, that's generally $1 billion/$2
22   billion/$3 billion, but I've also heard $10 million and $25
23   million.
24             David, I think you made a good point there could be
25   a huge difference, factors of 100, in the way we treat
0083
 1   companies.  
 2             Are folks thinking about small cap and emerging
 3   markets the same way?  Are we looking at $10 million
 4   companies, $25 million companies, $100 million, or are we
 5   still in the $1 billion and $2 billion companies?  
 6             Should all those be treated the same or do we need
 7   to have different regimes for each of those?
 8             MR. SOLOMON:  I have to say from our standpoint, I
 9   view everything $1 billion and south is small, small or
10   micro.  I don't really draw a distinction between small and
11   micro.  I feel like they are all the same.
12             I'm not really for having all these different
13   regimes for different sizes.  I just think fewer choices is
14   better in this particular case.
15             Once you get to $1 billion, there are enough people
16   to pay attention and that's not really an issue.
17             If you look at where research coverage has as a
18   proxy for where liquidity is, you will see if you're $1
19   billion and above, there are plenty of people that cover you,
20   and you don't really have issues in terms of getting access
21   to information in the market, and the liquidity improves
22   significantly at $1 billion and north.
23             I think from my standpoint, I would say $1 billion
24   and south is really where we are focused.
25             Just to kind of give you guys an idea, biotech has
0084
 1   been around for two decades or so, a little more than two
 2   decades maybe.  I just want to point out that there are still
 3   only about 20 companies that have market caps north of $1
 4   billion.
 5             All of the financing that has gone on, all the drug
 6   discovery and all the creation of -- what I would say is in a
 7   very robust capital raising part of the economy, we still
 8   have the vast majority of those companies sub-$1 billion
 9   period.
10             The expense associated with researching those
11   companies is not cheap.  It's expensive.  They're
12   complicated.
13             We can boil it down, rather than just look at the
14   numbers, look at specific industry types.  
15             Just in terms of the micro cap, Microsoft only
16   raised like $56 million when it went public, something like
17   that.   There were 114 investment banks and the spreads were
18   still at five or six percent.
19             When you actually think about that and you think
20   about the fact that last year Linkedin went public and they
21   raised $350 million and there were four banks on the cover.
22             Ask yourself the question, how does that happen. 
23   If spreads stayed constant, why did Linkedin -- how could
24   they only have four banks, and the answer is those banks need
25   that economics from the IPO to justify research coverage.
0085
 1             Ask any large organization, you look at the way the
 2   covers read today, the large banks are commanding 70 and 80
 3   percent of the economics in the IPO because they know they
 4   can't make any money in the after market.
 5             I don't particularly think that's healthy for a lot
 6   of companies that don't raise to the level of the attention
 7   of a Goldman Sachs or J.P. Morgan.  It's not healthy.  It's
 8   not a healthy environment for them.  
 9             We should have a number of different investment
10   banks.  The way to create that is to create economics that
11   are balanced and reasonable in the after market so people can
12   make money either on a net basis or a commission basis.  We
13   should have that.  It should be a stated objective of this
14   body to try to foster that.
15             MR. WEILD:  Gregg, I can send it to you.  We use $2
16   million as small cap and small micro cap, sub-$500 million
17   nano-cap, sub-$100 million.
18             What troubles me,and I would agree with everything
19   that Jeff said, except I'd take it even higher because we
20   have a sample set that we look at, IPO performance trends
21   over time for the large IPO, defined as larger than $500
22   million.  This is initial float size.
23             That has been declining and converging on success
24   rates for smaller IPOs as time has gone on.
25             I had conversations with -- he's now over at
0086
 1   General Atlantic -- David Topp, who used to be the head of
 2   Global Equity Capital Markets at J.P. Morgan.  His view was
 3   you need at least half a billion dollars in float to be able
 4   to sustain organically after market interest in the current
 5   market structure.
 6             MR. NALLENGARA:  We are short on time.  Cromwell?
 7             MR. COULSON:  $500 million.  Below $500 million,
 8   your shareholder base is going to be all stock pickers for
 9   the most part.
10             MR. NALLENGARA:  Josh?
11             MR. GREEN:  Just very quickly from the venture
12   capital perspective.  We really look at 300 to $400 million
13   thresholds because of the way the venture capital investment
14   model works to be the logical floor for a market cap to go
15   public.  That is what our industry, our companies are focused
16   on.  Anything below that would be a micro cap. 
17             We also have to account for the fact that after
18   market trading, it may take those numbers lower than those
19   market cap numbers, and we still need to support those
20   companies because they can dip and then come right back up as
21   well.
22             MR. NALLENGARA:  I hate to cut this conversation
23   off because it has been fascinating and extremely helpful for
24   us.
25             As I scanned through the questions, I think we have
0087
 1   covered the topics.  
 2             I want to thank the panelists for participating. 
 3   Hopefully, they will stay around, if anyone out there wants
 4   to corner them and pepper them with more questions, I am sure
 5   they won't like that I'm going to say they would welcome
 6   that.
 7             We are going to take ten minutes and then we will
 8   be right back for our second panel.  Thank you.
 9             (Brief recess.) 
10   PANEL 2:  EVALUATING CONCERNS RELATING TO TICK SIZE FOR
11             SECURITIES MARKET GENERALLY
12             MR. BURNS:  We are going to try to get started
13   again.  Thanks, everyone.  This is a group that doesn't get
14   to see each other very often, I take it.
15             MR. NALLENGARA:  We are going to start now so if I
16   could ask everybody to take a seat.  If you want to chat,
17   there is a beautiful atrium there, you can do it there as
18   well.  
19             MR. BURNS:  Suitable for capital formation outside. 
20   Thank you.  Thanks, Lona.  He speaks with that authoritative
21   Canadian voice that demands respect.
22             We are delighted to be starting on our second
23   panel.  This group has a rather hard act to follow.  That was
24   a very brisk dialogue that opened up a number of lines of
25   inquiry.  We know we have brought together a really good
0088
 1   thoughtful group of individuals who will carry the dialogue
 2   forward.
 3             To set sort of an initial theme, we can begin to
 4   delve more deeply into the question of fostering the
 5   ecosystem for the secondary market.
 6             It could have been noted at the outset, this entire
 7   subject has taken up a notable amount of interest since last
 8   year.  I think I counted Senate and House hearings, there
 9   were three in July, September, December, where
10   representatives from a number of the institutions who are
11   here today talked about re-thinking the one size fits all
12   approach that comes with decimalization.
13             I am sure we are going to invite the panelists to
14   introduce themselves.  We will pose a question or two.  We do
15   have a number of fruitful areas that we can follow up on,
16   including the premise of a tick size as the issue relative to
17   things like other rules that govern our markets, Reg ATS and
18   order handling, which came up.
19             Thinking about how we benchmark and measure the
20   impact if we were to pursue a tick size pilot, the standards
21   of increased liquidity, the fostering of research.  A number
22   of fruitful areas we can explore as the panel goes on.  
23             I will turn to this group and maybe moving from the
24   left side of the table to the right, I'll invite you to
25   introduce yourselves.
0089
 1             MR. REED:  Thank you for that.  My name is Adam
 2   Reed.  I am Associate Professor of Finance at UNC, the Julian
 3   Price Scholar there.  
 4             My research has been on a wide range of trading
 5   issues.  I've done some work on mutual funds, some work on
 6   market participation.  Sort of a focus of my work has been on
 7   short selling and in particular the impact of the regulatory
 8   environment and the information environment with respect to
 9   short selling.
10             Also, I've done some work on IPOs, both from the
11   perspective of short selling, its involvement in IPOs, and
12   also from the perspective of what makes a firm choose to go
13   public in the first place.
14             MR. JIGANTI:  Good morning.  I first would like to
15   thank the SEC Commissioners and staff for their continued
16   hard work creating and maintaining the most robust financial
17   markets in the world.
18             My name is Paul Jiganti and I'm a Managing Director
19   at TD Ameritrade in charge of market structure and customer
20   advocacy.
21             I started trading in 1985 on the Floor of the CBOE
22   as an independent trader.  In 1992, I moved to Susquehanna
23   and traded for 16 years, doing just about every type of
24   trading that Susquehanna did during that period of time.
25             I was involved at the CBOE as a member or the chair
0090
 1   of many committees.  In 2007/2008, I was on the Board of
 2   Directors.
 3             Since then, I created a dispersion trading model at
 4   Geneva Partners, left to work as a consultant at NASDAQ,
 5   introducing a new index product line.
 6             In July 2012, I joined TD Ameritrade in my current
 7   position. 
 8             I applaud the SEC for continuing to explore ways to
 9   make the equity markets better for all investors.  Retail
10   customers have never had a better experience in the markets. 
11   They have the lowest commissions ever, narrower spreads,
12   firms with an array of routing destinations to choose from
13   for best execution, and protections, all in under a second in
14   most instances.
15             TD Ameritrade clients are filled at the quoted
16   price or better 96 percent of the time for the entire route
17   of the order.  My fear is any change to this fragile
18   ecosystem could be to the detriment of the small investor. 
19   Thank you.
20             MR. ISAACSON:  Good morning.  My name is Chris
21   Isaacson.  I'm the Chief Operating Officer of BATS Global
22   Markets, the third largest Stock Exchange here in the U.S.
23   and the number one equity market center in Europe.  I'm also
24   a founding employee of BATS back in 2005.
25             Good morning and thank you for the opportunity to
0091
 1   participate in today's Roundtable.  I think it is a very,
 2   very important topic that is being discussed.
 3             Decimalization has led to tighter bid/offer spreads
 4   resulting in reducing costs to investors that Paul just
 5   talked about, but as we heard in the first panel, one size
 6   fits all approach to tick sizes may have unintended
 7   consequences of impeding the capital raising ability of small
 8   and mid-cap companies.
 9             Tick size regimes can also increase the cost of
10   trading in large, mid and small cap companies alike,
11   depending upon a company's liquidity profile and share price.
12             As a result, in Europe, for instance, the industry
13   developed a dynamic tick size regime designed to find an
14   optimal balance between incenting liquidity at the inside
15   while minimizing the maximum spread investors would pay to
16   access that liquidity.
17             We are proud to have played a leading role in that
18   initiative, which has been successfully adopted across
19   Europe.
20             We are encouraged that in hosting this Roundtable,
21   the Commission is examining not only concerns regarding
22   potentially larger minimum tick sizes for small and mid-cap
23   companies, but also broader questions about the current tick
24   size regime.
25             We believe that a wider tick size may make sense
0092
 1   for new issuers or illiquid securities, small and mid-cap
 2   companies, as well as certain higher priced but still very
 3   liquid securities, where an one cent price change may be
 4   immaterial to the notional cost of execution, but where
 5   investors are forced to access multiple pools of liquidity to
 6   complete a single trade.
 7             We also believe that a narrower tick size is
 8   appropriate for certain highly liquid lower priced securities
 9   where evidence suggests that an one cent minimum tick size
10   may be constraining price formation.
11             We also believe that as part of any pilot program
12   analyzing effects of changes in the minimum tick size the
13   Commission should consider allowing trading centers to
14   display locked markets in certain highly liquid securities.
15             Because of the ban on locked markets, most
16   Exchanges have implemented their systems such that better
17   priced orders are frequently displayed to reflect the
18   compliant yet less aggressive price level.
19             This is known in the industry as "price sliding"
20   and is necessary under Reg NMS to prevent locking the
21   markets.  Even though these orders would have improved the
22   national next bid or offer.
23             This in turn may be inhibiting the public price
24   discovery process and depriving investors of the benefits of
25   trading in a zero spread market.
0093
 1             BATS believes the Commission should implement a 12
 2   month pilot program with minimum tick sizes based upon
 3   considerations of the securities' quoted effective spread
 4   leeway, which is the number of ticks inside the average
 5   quoted spread, share price, and liquidity.
 6             Details of our proposal can be found in our written
 7   submission to the Commission, but in short, we are proposing
 8   a pilot program that would allow a five cent minimum tick
 9   size and an one cent minimum trading increment where either
10   the average daily volume in a security is less than 100,000
11   shares or the security is priced greater than $100 per share,
12   and the average effective spread that investors are realizing
13   as greater than five cents.
14             Secondly, a half cent minimum tick size in certain
15   very liquid securities with an effective spread of less than
16   one cent, and an average quoted size at the national best bid
17   or offer of 10,000 shares.
18             Finally, the display of locked markets in very
19   liquid securities with narrow effective spreads.
20             We believe this would be the most effective manner
21   for the Commission and the industry to determine the
22   potential benefits of tick sizes to companies and the
23   investing public.
24             Thanks again for inviting me to participate.
25             MS. O'HARA:  I am Maureen O'Hara.  I am the Finance
0094
 1   Professor at Cornell University.  I do research on market
 2   micro structure.  I also am Chairman of the Board of ITG.  We
 3   are an agency/broker-dealer firm that does a lot of work on
 4   transaction cost analysis among other things.
 5             MR. HEALY:  Hi.  My name is Pat Healy.  I have an
 6   MBA and CPA, former CFO in the banking industry, and for the
 7   last 17 years or so, I have run a company called Issuer
 8   Advisory Group.  As a result of that, I bring to you today an
 9   unique perspective, and that is of issuers.
10             I can tell you without getting into a long
11   description here that issuers generally like decimalization a
12   lot.  However, there have been some unintended consequences,
13   and the general consensus of the issuer community is their
14   shares, which are traded on these markets, largely become
15   chips in a casino.  
16             They would like to see some changes, and I think
17   they would wholeheartedly favor some move toward the topic of
18   discussion here today.  Thank you.
19             MR. HATHEWAY:  Thank you.  I'm Frank Hatheway.  I'm
20   the Chief Economist to NASDAQ Stock Market.  
21             I would like to thank the staff for including me in
22   today's proceedings, and we will talk about erasing the tape
23   in a minute.
24             Prior to joining NASDAQ, I was a finance professor
25   at Penn State University.  Prior to that, I was an options
0095
 1   trader in the old days.  
 2             In my role at NASDAQ, I talk to regulators, members
 3   of the trading community, listed companies, large and small,
 4   about issues of concern to them, market structure is a
 5   particular focus of what I spend my time on and my group.
 6             I arrived at NASDAQ shortly after decimalization as
 7   we were finishing our report on decimalization.  To summarize
 8   some of the less well covered parts of that report, small
 9   stocks did not particularly benefit, a quarter of the stocks
10   listed on NASDAQ, most of them inactive, actually experienced
11   an increase in spreads with decimalization.
12             Since then, spreads have tightened for all
13   securities, as Maureen said in her review of the trading cost
14   analysis work ITG has done, particularly for larger stocks,
15   many of which are tick constrained, trading at a penny
16   increment.
17             For the less active stocks, the benefits have been
18   less obvious and probably less dramatic.
19             When you think about liquidity in smaller
20   companies, a pilot to see if perhaps wider quote increment
21   might improve the environment for these companies, I think
22   it's a good idea.
23             More generally, tick size is just one element of
24   market structure that is overdue for assessment.  NASDAQ has
25   reported multiple Commission proposals in improving equity
0096
 1   market structure and moving us away from an one size fits all
 2   approach in many, many aspects of how we run the markets.
 3             The markets are continuing to evolve.  Regulation
 4   must be an iterative and incremental process.
 5             I encourage more dialogue such as this one to
 6   identify and evaluate alternatives which improve the market
 7   for investors, listed companies, and the American public.
 8             I look forward to your questions, and thank you
 9   again.
10             MR. CONCANNON:  I'm Chris Concannon, Partner and
11   Executive Vice President of Virtu Financial.  We are an
12   electronic market maker.  We make markets in several thousand
13   companies around the globe, here in the U.S., Canada, Europe
14   and Asia.
15             We also run a designated market maker on the Floor
16   of the NYSE where we employ actually human traders, and we
17   make markets in about 250 names which are by definition small
18   and micro cap stocks.
19             I think the most important point, some of us have
20   touched on the topic, decimalization of all the market
21   structure changes that we have witnessed over the last 20
22   years, decimalization has delivered more profits to investors
23   than any other market structure change.
24             In the context of this Roundtable and talking about
25   decimalization, it was an unbelievable success.  Where we
0097
 1   have failed our market is where we have treated it in the one
 2   size fits all, and we will hear that phrase over and over
 3   again.
 4             I do think we need to at a Federal level customize
 5   our market to the nature and characteristics of the
 6   instruments that we are trading.  That means trade increments
 7   need to be adjusted.  That means market structure features
 8   need to be thought about. 
 9             We come ironically from a market structure that did
10   in fact have different market structures for different
11   symbols.  If you think about the 1990s, despite all the
12   flaws, there was a dealer market in the OTC, and many
13   companies started on the dealer market and then they
14   graduated to the auction market, either on the American Stock
15   Exchange, and if they were very successful, they would
16   graduate to the NYSE.
17             We did in fact accidentally have a migrating market
18   structure that was dealing carefully with the symbols that
19   were listed on those markets, and today, listing standards
20   are currently across our markets which are very low.  We have
21   some of the lowest listing standards we have ever had.  
22             All companies just fall into the same market
23   structure that a Microsoft and Cisco exist.  I think that is
24   ultimately the flaw that we are identifying on this panel,
25   not questioning the success of decimalization.
0098
 1             Thank you for having me.
 2             MR. CRONIN:  Thanks for having me.  I appreciate
 3   the opportunity to speak here today.  My name is Kevin
 4   Cronin.  I am from Invesco.  Invesco is a global independent
 5   asset management firm with assets under management of $687
 6   billion.
 7             Essentially, I am here to represent the interests
 8   of long term investors, which I think all would agree are the
 9   cornerstone of efficient and effective market structure and
10   capital formation.
11             Within the context of that, I will tell you that I
12   think generally we do believe there have been advances made
13   with respect to our investors' ultimate outcome as a result
14   of the changes in market structure.
15             I think some will take probably more credit than
16   they are due on this panel.  I think actually if you were to
17   dis-aggregate our interest in things, in terms of transaction
18   costs, as we will get into invariably, spreads is not one of
19   them.
20             Narrowing spreads is an interesting academic
21   concept but in the construct of an institutional trade, when
22   you have 150 or 200 percent of the daily volume, spread
23   doesn't really factor into the equation.
24             We worry instead about things like posted
25   liquidity, not transient posted liquidity, it's really there.
0099
 1             We think while there have been tremendous strides
 2   made in market structure, there is clearly room for
 3   improvement.  We certainly applaud the Commission for
 4   examining decimals.  
 5             I think as others have described, it certainly
 6   sounds reasonable to take a look at how decimals have
 7   interacted in the marketplace and the impact, intended and
 8   other, that has been had as a consequence.
 9             Our own theory is that certainly from a large cap
10   perspective, things are better, but it's far less clear to us
11   as others in the mid-cap and small cap space that the
12   benefits of electronic trading, algorithmic trading and other
13   things that we now take for granted have really brought the
14   level of sophistication, equality, efficiency to the mid and
15   small cap world that I think most of us would have hoped.
16             I thank you again.  I would like to again take the
17   opportunity, I suppose I would be remiss, as the only
18   institutional investor on the panel, to say while we
19   certainly appreciate the investigation of decimals, there are
20   other issues that we would hope regulators would take a look
21   at in the not too distant future, certainly things like
22   maker/taker economics, internalization, the proliferation of
23   order types and the lack of a consistent regime around the
24   disclosure of those order types, and lastly, order
25   cancellation, something we probably all see a lot as we see
0100
 1   flickers and that sort of thing.
 2             As far as I can see, it doesn't add to investor
 3   confidence and quality of markets.  Certainly, to confusion
 4   and at some point probably deleterious to investor
 5   confidence.
 6             Thank you again and I look forward to answering
 7   your questions.
 8             MR. BURNS:  Thanks to all the panelists for that
 9   introduction.  Thanks to Frank for that moment in levity. 
10   Another merger.  We were all kind of wondering.
11             Before I turn it over to my colleagues in Trading
12   Markets and Risk Fin and Corp Fin for questions they might
13   ask, I think maybe just as a table setting exercise, why
14   don't we jump right into measures, liquidity, fostering
15   research.
16             The European models are going to be an interesting
17   topic.  
18             Is there anything -- we began to talk about the
19   impact of decimalization in the last panel, is it a cause, is
20   it a symptom, is it a consequence or something coincident or
21   additive.
22             Anything you would like to augment, touch upon,
23   anything you think we didn't get into?  Anything else?
24             MR. CONCANNON:  I would just add -- first of all,
25   we are supportive of a pilot.  I think we are not going to
0101
 1   know -- there are no studies that will really give us a
 2   conclusion around trade increments in the U.S.  There are so
 3   many different market structure features of our market versus
 4   other markets.
 5             I think they are wonderful guides.  The AMS study
 6   was actually very impressive.  I think there are a number of
 7   things that suggest we should do something.  The one thing I
 8   would suggest is in the context of a pilot program, I would
 9   take a hard look at market maker obligations.
10             If we are going to change the trade increment, we
11   should take a hard look at market maker obligations, with the
12   interest of increasing those obligations, in exchange for an
13   increase in the potential profit as a result of the trade
14   increment.
15             MR. BURNS:  I'm going to turn to my colleague,
16   Gregg Berman, from our new Office of Analytics and Research.
17             MR. BERMAN:  I would like to ask all the panelists
18   their opinion on why the market hasn't solved this aspect or
19   this problem for itself.  
20             Certainly, Commission rules are consistent with a
21   marketplace aim we would like to have, the two cent or five
22   cent tick, and competitive pressures may say if I do five
23   cent but someone else has an one cent, that's a problem.
24             We are hearing from everybody saying yes, this is a
25   problem.  What's preventing the marketplace from saying
0102
 1   actually we can have our own pilot, so the need for a
 2   specific Commission action on this.  
 3             I'm wondering if folks can comment on that.
 4             MR. HATHEWAY:  I think you can get some insight on
 5   that on what's happened even where we have a tick and it's
 6   appropriate with a penny.  We don't have uniformity because
 7   there are very, very strong economic incentives to defect
 8   from what everyone else has to do.
 9             Everybody follows the rule.  There are exceptions
10   to the rule.  If you can exploit those exceptions, you do so. 
11   It gives you an advantage.
12             In the European context or here in the United
13   States, if you have a voluntary approach or you have an
14   approach that is taken unilaterally, and we have tried some
15   with different market structures, it isn't necessarily going
16   to produce the outcome you desire, either experimentally,
17   speaking as an economist, or practically speaking as a
18   representative of NASDAQ, because there are always
19   opportunities for people to do what again with the academic
20   hat on we would call an epsilon better, just a tiny bit
21   inside what you are trying to do.
22             I think that is the main reason you haven't seen a
23   market solution, that this is a situation where you probably
24   have a legitimate market failure around coordination.
25             MR. CONCANNON:  I would agree.  I think the
0103
 1   challenge -- we chose a path which is a competitive market
 2   structure, and that means those Exchanges and those
 3   alternatives are going to compete with one another.
 4             The outcome of that is actually very helpful for
 5   the economics of the end user, but it also delivers this
 6   issue where sometimes the Federal Government has to say hey,
 7   we all need to agree on these certain dynamics, so maybe it's
 8   a national market system.
 9             I think at the national market system level, they
10   could do it, but they do need a push, a fairly strong push to
11   decide on trade economics.
12             You can't have one Exchange have different quote
13   increments than other Exchanges.  It just won't work.  That
14   is what you had in Europe.  They tried to have an industry
15   solution.  It's been helpful.  There clearly needs to be a
16   broader solution in Europe as well.
17             MR. ISAACSON:  I will speak to the push in Europe
18   that we were part of.  There was somewhat of a race to the
19   bottom regarding tick sizes, and we came together and said
20   this is not in the interest of anyone, we should agree upon
21   tick size tables, and while not completely uniform across
22   Europe, I think where we landed is a rational tick size
23   regime.
24             It always can be improved, but there is an ongoing
25   evaluation of those tick sizes and what securities fall into
0104
 1   the tick sizes, and whether or not they should be in or out. 
 2   I think that is a healthy evaluation if given enough time to
 3   gather data about how the security is or is not trading.
 4             I would also say that I did sign a letter a couple
 5   of years ago, our timing was impeccably poor, before the
 6   flash crash on May 6, with NYSE and NASDAQ saying let's do a
 7   tick size pilot.  In that case, it was regarding tighter tick
 8   sizes for very liquid securities, which we have yet to hear
 9   from the Commission on.
10             I would say we have come together in that regard. 
11   We would still stand behind that but the reason we are here
12   today is about wider tick sizes for small and mid-cap
13   companies.  I think for the same reasons we could come
14   together, but I don't think one of us doing it is going to be
15   good for the industry.
16             Similarly, lessons learned from the flash crash,
17   one thing was volatility mechanisms implemented by a single
18   Exchange in a fragmented market doesn't make much sense.  
19             Tick size experiments in a fragmented market by a
20   single market doesn't make much sense.
21             MR. HEALY:  To broaden your question just a little
22   bit.  You talk about market solutions, and typically when you
23   talk about market solutions, it's the trading community.
24             I think the issuer community would come at that
25   from a different angle.  They would say why don't you let me
0105
 1   pick the tick size.  I'm the guy who is sitting there with no
 2   liquidity in my stock.  I'm the guy that can't get anybody to
 3   cover me.  Why don't you give me the option to go to a
 4   nickel, and if I think that's good for my shareholders, I'll
 5   do it.
 6             You have to make that available to get the market
 7   solution that you are talking about.  That is not currently
 8   available.
 9             MR. REED:  I would just add echoing Chris' thought
10   a little bit.  The extent to which this is still seen as a
11   problem really depends on the point of view and who you are
12   asking.  
13             I think there has been tremendous benefit to
14   decimalization, and I think this issue is inherently
15   controversial in the sense that tick sizes benefit some
16   market participants at the expense of others.
17             I think when you are sort of thinking about this, I
18   absolutely applaud the Commission for taking a careful view
19   on this and think about how the data has been  used in the
20   past, how the data reflects the views of those people making
21   the points.
22             MR. CRONIN:  My own sense has been that the
23   conflicted economic models that exist today certainly don't
24   perpetuate a solution in the industry, so Exchanges becoming
25   for profit, for example.
0106
 1             We have seen a number of different competitive
 2   efforts but am I the only one that seems to sense that most
 3   of that is around pricing.  
 4             We hope that the competition that would be spurred
 5   by NMS would really be about innovation and ways that we
 6   could make trading costs less impactful.
 7             My sense is unless and until there is either a
 8   mandate that it goes to a specific level or some greater
 9   change in the economic incentives of all, that it is going to
10   be very, very difficult to get to a solution that I think is
11   amenable to all.
12             The other thing I would say is while I recognize
13   that there are going to be those with different thoughts
14   about how this whole thing has kind of played out from their
15   own perspectives, I would take a step back and say as we try
16   to prioritize, what's the important voices in this whole
17   thing.
18             I do hope that we don't lose sight of the fact that
19   retail investors by and large get the majority of their
20   exposure from institutional investors.  As we try to solve
21   this complicated problem, let's not just address success by a
22   self directed retail investor.  
23             Again, if we are solving it for the greater of
24   market participants, we would just suggest that institutional
25   interests are obviously very important as well.
0107
 1             Unfortunately, there is just a natural tension that
 2   exists between what might be the best outcome for retail
 3   oriented investors, the self directed variety, and bigger
 4   institutions.
 5             In recognition of that, when we talk about things
 6   like dark pools and other things, there are real reasons to
 7   kind of keep things different, that this one size fits all
 8   paradigm just doesn't work in the U.S. market structure nor
 9   should it.
10             I think that applies not only to retail and
11   institutional phenomenon but clearly to capitalization of
12   companies and stocks and how they trade.
13             MR. JIGANTI:  You can't have the retail investor --
14   you have to have them present at the table when these things
15   are thought of.
16             I believe that as retail investors look at the
17   marketplace, that's the advertisement.  That's the part of
18   the market that is at the inside quote.  The institutional
19   investors have every tool at their advantage to use all the
20   different market mechanisms to achieve their desired
21   position.
22             MR. CRONIN:  Of course, you recognize that if you
23   are buying 200 shares, it's a far different proposition than
24   buying 300 percent of the daily volume.
25             We have to have additional tools, obviously,
0108
 1   because of the impact that our orders could have in a
 2   marketplace.
 3             I would say that these kinds of tools should be
 4   available to all kinds of investors, and by my best
 5   recollection, there are a number of different ways that
 6   retail investors get exposure today and access to the markets
 7   that they clearly didn't have.
 8             As we talk about advances in market structure, it's
 9   clear that there have been real advances made.  I think the
10   retail investor in particular has been a tremendous
11   benefactor.  That is great.
12             Let's make sure -- we are sorry for this
13   complicated problem -- that it is not one dimension we are
14   trying to solve for.  That's my only point.
15             MR. JIGANTI:  I guess I'm looking at the market as
16   a whole, in the lower cap stocks.  When you are talking 300
17   percent of the average daily volume, that is completely
18   different.  
19             In our comment letter about this, if we are going
20   to go forward with a pilot program, it does make a lot of
21   sense in some of these areas to widen the tick size.
22             My concern clearly is if you are solving for
23   liquidity, it's one thing.  If you are solving to get more
24   research in that group, it's completely separate.
25             MR. CRONIN:  I think as we have looked at this,
0109
 1   this isn't really about necessarily solving the research
 2   problem.  It is about solving the liquidity problem.  
 3             In many respects, that becomes a virtuous cycle. 
 4   Why don't we post bids and offers today in small and mid-cap
 5   stocks?  Because the price increment that somebody can jump
 6   in front of those posted bids or offers is a penny.  It's a
 7   very minimal amount of risk that is required to take, to jump
 8   in front of an order.
 9             What we are saying is well, in large cap, that's a
10   problem, so we just hide more things, we break it up over the
11   course of the day, we adapt and we adjust.
12             In the small and mid-cap space,  you just don't
13   have that secondary liquidity part that happens with natural
14   market makers or whatever, because not only is it then the
15   erosion of an incentive, frankly, from an investor's
16   standpoint, but so, too, from a market maker's standpoint, if
17   you can't really make any reasonable economic return on the
18   investment, you won't bother making markets.
19             I think the proposition that we have put forth, and
20   again, this is not necessarily supportive or not of research,
21   is this is a liquidity matter.  If you can solve the
22   liquidity matter, nothing draws a crowd like a crowd.
23             I suspect that if you were to find there were more
24   incentives to post bids and offers for institutions, as
25   institutions view the stocks that typically wouldn't fall
0110
 1   within their screens because of liquidity concerns, trading
 2   just doesn't support what's necessary for a mutual fund to
 3   buy a security, if you change that paradigm, all of a sudden
 4   you have changed the attraction of that security, which by
 5   its own right brings more trading activity to it, and I
 6   think, again virtuous things can happen.
 7             The question isn't would it be reasonable to take a
 8   look at this because decimals have been in place for 10 or 12
 9   years.  The question is is there a way to actually make this
10   much more effective for a group of stocks that don't seem to
11   have participated at the level of benefit that some of the
12   others have in the current market structure.
13             MS. EDWARDS:  I have a quick follow up question for
14   Patrick.  There is a stream of economic literature that says
15   issuers do have control over their tick size, and that is
16   relative tick size relative to their price.  They can split
17   their stock and basically affect their tick size.
18             Why doesn't that work?
19             MR. HEALY:  I don't really think we are talking
20   about the right part of the market that is in a position to
21   split their stock.  This is the low end, the low cap.
22             I think if you look at those statistics, I'm not
23   sure those are even splittable stocks.
24             What we are talking about here is thinly traded
25   stocks that can't get coverage.  Split the stock, put more
0111
 1   shares out there as a percentage of volume, it will increase
 2   the liquidity in the stock.
 3                  I'll go back to something that Maureen said in
 4   the prior panel, and that is you should be focused on the
 5   dollar value traded.  You can split the stock all you want. 
 6   It is not going to increase the dollar value traded at all. 
 7   It's going to be more shares at a lower price.
 8             I don't think that is getting to the issue.  The
 9   issue is the economics have been drained out of the research
10   side.  That CFO is sitting there and he would love to get
11   some research.  He would be happy to pay for it.  If he were
12   to pay for that research, it would be tainted by the payment.
13             The only way for him to get the exposure he's
14   talking about here is for the economics to return to that
15   part of the research business.
16             Splitting the stock gets you a couple of headlines,
17   gets you a little talk, gets you more shares outstanding.  It
18   doesn't do anything with the dollar value traded and it
19   doesn't do anything in terms of getting your research
20   coverage.  That, I think, is the problem.
21             MR. HATHEWAY:  I've had these conversations with
22   companies large and small.  The answers are several.  It is
23   costly, one of the big ones.  The main one for smaller
24   companies, say $20 stock that wants to get the equivalent of
25   a nickel, it's a $4 company, once it does the split.  $4, can
0112
 1   you guys hold stocks much below $4?
 2             MR. CRONIN:  We hope they go a lot higher.
 3             MR. HATHEWAY:  Yes.  Firms begin to actually become
 4   concerned about meeting listing standards/thresholds if they
 5   get their stock price too low.  I think for the small
 6   companies, those are issues.  
 7             Another thing that often gets ignored, if you are
 8   talking about a high multiple split for a higher priced
 9   stock, you have to buy back fractional shares typically.  We
10   are talking about companies that are not big and are not
11   flush with cash and not in an easy position to buy back
12   fractional shares or fractions of round lots.
13             There are a number of quite practical economic
14   reasons that people don't manage their stock price to produce
15   desired tick size.
16             MR. BERMAN:  Is there a reason why they don't start
17   off with the tick size?  I had an opportunity to take a
18   company public.  The number you are looking for is 20, give
19   or take 2.  There is something about a $20 stock.  Doesn't
20   matter how big or how small.  That seems to appeal to
21   investors, both in retail and institutional.
22             If I wanted to have a tick size that was ten times
23   larger, I would just go out at $2 to start with, really
24   doesn't change any of the economics of the company at all,
25   but I have created a tick size by myself without any action
0113
 1   on anybody's part, that is effectively ten times larger.
 2             In Europe, stocks trade in the single digits all
 3   the time.  It seems to be very natural that stocks trade at
 4   50 pence, 80 pence, 1.5 Euro.
 5             What is it about the magic numbers that it has to
 6   be in the mid-two digits, apart from listing standards, but
 7   presumably all those were geared -- based on the same
 8   results.
 9             MR. HEALY:  I think I can help you with that.  Put
10   yourself in the seat of a CFO doing an IPO.  An investment
11   banker comes over and he says look, $20.  You're thinking
12   maybe $10.  All of a sudden, you're a Groupon or another tech
13   stock that has seen a 75 percent decline in your market value
14   because you came out of the gate very strong and things
15   didn't go well.
16             All of a sudden, you're a penny stock.  You have
17   listing issues.  Nobody is going to take that risk.  It's
18   really a matter of risk management.  $20 is a very prudent
19   level to price the stock to give yourself some room.  
20             If you look over the last couple of years,
21   particularly in the tech sector, room has been necessary. 
22   That's why they do it.
23             MS. O'HARA:  I'd like to just add a couple of
24   comments about these issues of quoting increments and
25   liquidity.
0114
 1             Again, it's kind of an interesting question about
 2   we do know that stocks with low prices have higher relative
 3   ticks.  It isn't true that in some sense, we have the same
 4   tick size because it differs.  
 5             An interesting way to think about these issues is
 6   to think about matching stocks on industry and market cap
 7   that differ with respect to the price levels they trade at.
 8             I've been working on some research along these
 9   lines, looking at that kind of issue.  I'll point out we are
10   still not finished with this research, so I can't quite share
11   it all yet.  
12             One thing that is very intriguing is that stocks
13   that have higher relative ticks, professional market makers
14   spend more time at the tick and provide more orders than do
15   institutions.
16             It's kind of an interesting issue if you compare it
17   relative to your control group.  It incentivizes the
18   professional market makers and they have changed their
19   behavior more than the institutional traders.
20             Again, as we think about how changing ticks affects
21   the big picture, we have to recognize that for many stocks,
22   in fact, maybe for most, the bulk of the liquidity now is
23   provided by sort of the professional market makers.  
24             This can affect the liquidity. Whether it affects
25   the research that you are looking for, I'm not so sure.
0115
 1             MR. HEALY:  If I could respond to that, I think
 2   what you are saying is that the economics of that relative
 3   tick size gives rise to professional traders who add
 4   liquidity.  I think that proves my point.
 5             If you put the economics back into the equation,
 6   you are going to get more research coverage.  It's going to
 7   happen. That is why I think issuers are craving coverage
 8   because they want liquidity in their stocks.
 9             MR. CONCANNON:  Just two points.  We heard it on
10   the last panel and we are hearing it now.  When I think about
11   tick size and even a pilot, I think when it comes to
12   research, Wall Street has a research issue, regardless of
13   tick size.  It has a research issue.  If you talk to the
14   large banks writing research, they don't even think about the
15   small caps in this research issue.
16             They are having a hard time paying for research in
17   the large cap activity they have.
18             Wall Street has this research issue, tick size, we
19   could go to $1, we are not going to solve the research issue. 
20   That's how expensive research is.  
21             I also agree that companies benefit from research
22   and research coverage, but I think that is one of those
23   things the market has to solve.  
24             Playing with the market structure, micro market
25   structure to solve that dynamic, I think will ultimately be
0116
 1   too costly to the end investor, that we have to fund and
 2   create inefficiencies in our market to somehow create this
 3   research demand when it is a problem no matter what.
 4             I want to make sure it's clear that this is more
 5   about liquidity.  This pilot is really studying the posted
 6   liquidity on our market and whether or not we can adjust
 7   that, and if we do adjust it and we are successful, does it
 8   benefit the buy side and does it benefit the retail.
 9             That should be our narrow kind of goal of analysis. 
10   Whether or not we create research, we trade 250 names of
11   small caps, we don't make enough money to put people on the
12   Floor let alone write research.  
13             I can e-mail you a page from Yahoo.  That is about
14   all we can do.
15             MR. BURNS:  Let me pick up on a point, if I may,
16   going back to something Pat said, talking about issuers
17   determining tick sizes.
18             Two things.  I'm taking up something from the last
19   panel.  Is there a number we ought to be looking at on size
20   of company, above or below which, for purposes of the pilot,
21   we ought to be looking at?  I think a number came out at
22   about $1 billion, maybe someone said $500 million.
23             Secondly, how would an opt in work?  I'm thinking
24   of where we sit as regulators, thinking about a potential
25   pilot.  Can you really have an opt in regime, should you have
0117
 1   an opt out regime.  That is to say you try to come up with
 2   kind of a cross section.
 3             I'm thinking of your CFO as well, who has to go
 4   into the CEO and say guess what, boss, we're in this pilot,
 5   and you don't really know what the outcome is going to be.
 6             Sort of a compound question but throwing it out for
 7   you guys to talk about.
 8             MR. HEALY:  Let me take the first crack at that. 
 9   First of all, I think and I hope you will do the pilot.  As
10   you embark on that, I think we need to be careful in a couple
11   of ways.
12             One, we need to isolate the variables to the
13   problem at hand.  The quality of the company is every bit as
14   important as the tick size.  In fact, more important.  You
15   get to the question -- I think the essence of your question
16   is or one of the parts of the question is who do you include.
17             I would say and I am sure the fellows from the OTC
18   market would not support this, but I think first of all the
19   stock has to be listed on either NASDAQ or the New York Stock
20   Exchange because of the higher listing standards.
21             Secondly, I think it has to be profitable.  The
22   reason it has to be profitable is we are trying to remove
23   variables from this question and we want to focus on what is
24   the impact of tick size to liquidity.
25             It's very important that we get a control group
0118
 1   that doesn't have other factors that influence liquidity.
 2             I do think in response to your other question the
 3   opening bid on this thing ought to be a nickel, a nickel
 4   spread.  To add to that, I believe it should be issuer
 5   choice.  
 6             I think you will have a group of profitable NASDAQ
 7   and New York listed stocks who will line up.  You will have
 8   to have a Lotto to determine who is in this pilot.  
 9             The CFO is going to say look, I'm frustrated.  I
10   can't get our story on Wall Street, I don't have any research
11   coverage, we don't have any investment bank deals pending,
12   where else am I going to turn.  I'm going to volunteer for
13   the pilot so that maybe I can draw some attention to us.
14             I have a great story, nobody is listening.  
15             MS. O'HARA:  Can I just raise an issue?  This is
16   the academic in me.  Academic's hate things where people get
17   to sign up, because if we are actually going to be able to --
18   if we want to actually be able to ascertain the impact of
19   this experiment, we need to be able to control for everything
20   else.
21             When you allow people to sign up, there will be a
22   certain type of company that may find this more attractive
23   than others.  Now as we try to put together a control group,
24   you really run into problems.
25             Should the Commission do this, perhaps one way to
0119
 1   address your concerns is the initial pilot does not involve
 2   people volunteering, but having seen the outcome of the
 3   pilot, perhaps the long run down the road might be allowing
 4   that to happen.
 5             MR. HEALY:  I respect your opinion.  I just want to
 6   raise the point here that you are not going to have any
 7   shortage of volunteers.  You are going to have a volunteer
 8   Army that is more than you need.  You won't have to go to a
 9   draft here.  You will get issuers that participate.
10             MR. REED:  I would like to say I absolutely agree
11   that when academic's try to study these things, the word they
12   use is "endogenicity," that the type of issuer that would
13   choose a new tick size is somehow different from the type of
14   issuer who didn't choose a new tick size.
15             There might be sort of a hybrid approach where you
16   could allow an opt in but then somehow randomly assign which
17   firms actually got to participate in the pilot.
18             I think the Commission has shown excellent ideas
19   about this, the Reg SHO pilot.  There might be a hybrid
20   approach where you have randomization.  The question that the
21   academic's could answer would be slightly different in this
22   case, and it would be comparing opt in firms who made it into
23   the sample versus those who didn't make it into the sample,
24   and not able to compare those that didn't opt in, but
25   nevertheless, it would give the academic's a little something
0120
 1   in terms of an experiment that they could trust.
 2             MR. HEALY:  I would just tell you, the companies
 3   would defer to your good judgment on that.  That would not
 4   become an issue.
 5             MR. JIGANTI:  You would have to compare them
 6   against like companies in the market that we are looking at
 7   today.  If you're going to say it's an opt in program but
 8   we're going to compare it against the companies that wouldn't
 9   make it into your pilot, I think that is where the problem
10   is, and I think you came up with a great solution, Adam.
11             MR. HATHEWAY:  I just want to support what Pat is
12   saying.  I don't think you will have trouble with companies
13   opting in, and we have probably a quarter of the list, small
14   stocks to very big stocks, that trade at a dime or more. 
15   That is ten ticks.  In the old days, nobody traded at ten
16   ticks if there was any interest in the security at all.
17             I also don't want to lose sight of the fact that
18   with our petition with Chris' firm and the New York Stock
19   Exchange, my former or future employer, we have probably 300
20   or 400 stocks that are trading at a penny or less.  
21             I take partial issue with Maureen's point.  As the
22   price comes down and the spread approach is a penny, that's a
23   displayed spread.  The traded spread actually comes well
24   within a penny because you have increasing levels of price
25   improvement, increasing levels of mid-point trading, and
0121
 1   other types of activities that essentially show a constrained
 2   spread on the stock.
 3             That raises costs for investors just as much as the
 4   spread that is sort of too wide and too unstable.  
 5             We think a little bit about sort of the broader
 6   context here.
 7             If you are doing a pilot, you want to do a pilot
 8   that gives a little and takes a little.  Parts of the pilot
 9   are lower costs to investors for narrowing the spread has a
10   certain appeal in some constituencies that may take a dim
11   view of widening the spread.
12             MR. ISAACSON:  I want to follow up on what others
13   have said regarding the opt in.  As I mentioned our comment
14   letter in my opening comments, I think there should be some
15   gross criteria set in order to be able to opt in, your
16   security has to be of a certain security profile or liquidity
17   profile today, trading less than 100,000 shares ADV or
18   trading less than a certain notational value every day or
19   lower than a certain market cap.
20             I think there is maybe asymmetry potentially in
21   potential conflicts of interest with an issuer saying I want
22   to trade a very, very liquid security, some CFO saying I want
23   my security to trade at a nickel or dime or whatever, and
24   that would be born entirely by the retail investor,
25   especially as Frank said, currently that security is trading
0122
 1   at a spread of less than a penny or at a penny today.
 2             I think there is just gross criteria, 100,000
 3   shares a day or $250,000 a day, some sort of opt in, and then
 4   randomization about who is in the control group as part of
 5   the pilot would be in order here.
 6             MR. HEALY:  I'm a big fan of common sense.  There
 7   were two parameters I threw out there, has to be a listed
 8   stock and they have to be profitable.  If there are other
 9   criteria that folks think make sense, I'm all for it.
10             Again, we're trying to isolate for variables here
11   that give us the right answer with respect to liquidity.  To
12   the extent that you have criteria that eliminates those
13   variables, I agree.
14             MR. ISAACSON:  One of the criteria, certainly not
15   as it relates to mid and small cap companies, but you have
16   outlier securities that have been incredibly successful
17   companies, Google and Apple, that trade at effective spreads
18   that are 5/10/20 cents per share.
19             I'm not convinced that a penny tick is the
20   appropriate tick size for those securities.  
21             Securities that have prices that are greater than
22   100 shares, I think, should be part of the pilot even if they
23   happen to trade more than a certain de minimis amount of
24   trading every day.
25             MR. CRONIN:  I think we would tend to agree with
0123
 1   that, but I think it's probably important as we approach this
 2   pilot program to do it very incrementally.
 3             I think the Hippocratic Oath probably applies well
 4   here, "above all else, do no harm."
 5             In the large cap space, I think generally we all
 6   agree that the experience is much better than it has been.  I
 7   don't think any of us disagree that there is probably room
 8   for improvement in some dimensions as well.
 9             The point is as you begin this study, it is both
10   important to sort of identify a group specifically as opposed
11   to a broad swath of the market, and really see if there is
12   value in this enterprise.
13             I think probably it does extrapolate beyond if
14   truly there is value. 
15             If I could just quickly opine on whether the
16   company should be the determinant of that.  I suspect there
17   is probably an element of investors and Exchanges who should
18   probably also be involved in that, because there are a number
19   of variables that kind of matter when you look at liquidity.
20             Whether it is included in an Index, whether it
21   actually has a top ten holder's list of real investors, these
22   kinds of things ultimately have a pronounced impact on
23   liquidity profiles.
24             Just saying that a company is $1 billion and is in
25   the pool and they have opted in or opted out, it doesn't seem
0124
 1   like it is setting ourselves up for real scientific data.
 2             By the way, we are all for data in these kinds of
 3   enterprises.  I don't know about you but I don't really have
 4   a day in my recent past where I haven't seen a study that
 5   says this is true or this is false, et cetera.
 6             The best way to try to figure out if these kinds of
 7   initiatives are valuable is to get the data.  I agree with
 8   the academic side on this, which is look, if you have a bunch
 9   of guys just opting in because they think it is going to be a
10   better outcome for their security, it's kind of comical from
11   my perspective to think somebody thinks that if their stock
12   is now at a nickel as opposed to a penny, that somehow there
13   is more intrinsic value in that.
14             The other side of it is I think if we were just
15   trying to get to a place where we are bringing value, trying
16   to least complicate this, this seems to be the best approach
17   and having people with the appropriate amount of perspective
18   to help make those decisions and set reasonable criteria with
19   no opt out's.  That seems like the right way to do it.
20             MR. JIGANTI:  I think a volatility component in
21   this would make a big difference.  If you measured volatility
22   along with the notional changes every day, the overall
23   volume, along with the other criteria, and really had a think
24   tank or group of probably similar to this Roundtable to aid
25   the SEC in finding the right mix would come up with a pilot
0125
 1   program that would probably be most effective.
 2             MS. O'HARA:  Can I raise a couple of issues?  If
 3   you are thinking of designing a pilot, ideally you want to
 4   make sure your control group is sufficiently a control group,
 5   so that you can actually draw inferences from what you get.
 6             There are a couple of things that make sense. 
 7   Market cap is exogenous to a firm in a sense.  Ideally, what
 8   you want to be able to select on are things that are kind of
 9   exogenous, things like market cap makes sense.  Industry.  
10             There can be differences in the liquidity across
11   different stocks because they have different appeal to
12   different sorts of investors.
13             You may want to think about controls based on size. 
14   I think you would be hard pressed to argue that once you get
15   above $1 billion in market cap, you are really talking about
16   small companies.
17             I think there are really two stories here.  One is
18   the really big guys and one is the little guys.  If you are
19   really talking about the little guys, certainly in the
20   studies I'm familiar with, stocks that are really below about
21   $250 million in market cap are really very different even in
22   stocks between 250,000 and 500,000.
23             Once you get above $1 billion, they are really not
24   so little any more.
25             As I look at these issues, things like size as
0126
 1   measured by market cap are important.
 2             The challenge when you begin to put in trading
 3   volume is trading volume in turn is actually part of the
 4   liquidity of the stocks.  The reason you have the volume is
 5   in part because of the liquidity you have, so that becomes
 6   exogenous and that gets kind of messy.
 7             There are interesting things.  A few years ago we
 8   might have said if you're going to have a control sample, you
 9   should be matching New York with New York and NASDAQ with
10   NASDAQ.
11             Nowadays, because of the nature of trading and
12   where things go, those issues are less important but it still
13   becomes the case that as you begin to form your control
14   samples, it may not be that easy to find stocks that are kind
15   of in the same industry but are just bigger.  Some of these
16   stocks are really so tiny, finding controls is challenging.
17             I just raise the issue, any kind of strategy that
18   you look at, and volatility fits into this, because
19   volatility is also kind of an endogenous variable that
20   depends on liquidity as well, so selecting on that is a
21   difficult thing to do and I wouldn't advise it.
22             However, looking at things like trading volume and
23   volatility as the outcome of the change is surely something
24   you would want to do if you were doing this pilot.
25             MR. HEALY:  If I could just add to that.  I do a
0127
 1   lot of comparisons between companies looking to either list
 2   on the New York Stock Exchange or NASDAQ.  Sometimes I've had
 3   companies looking to switch markets.  Sometimes I had
 4   companies in an IPO, and they want to see where others have
 5   been.  In other instances, you have spin off's and mergers
 6   and other things.
 7             Invariably, we come back to exactly what Maureen
 8   just said, and that is how do you get an apples to apples
 9   comparison.  We use multiple variables to come up with those
10   apples to apples comparisons.
11             For example, if you are in the same industry and
12   you are Kraft, there is only a handful of companies that you
13   can compare yourself to.  You can have trading peers.  You
14   can have companies of similar volatility, similar market cap,
15   similar stock price.
16             What you do is you find those common denominators
17   and you put them side by side and you say okay, is this
18   giving me the right answer.  
19             I really think that is what you are talking about
20   here as you tailor your list of companies to include in the
21   pilot, there are multiple factors that you want to take into
22   consideration.
23             Just to reiterate a point I made before, the
24   listing standards between NASDAQ and the New York Stock
25   Exchange are pretty darn close.  They do set a floor, at
0128
 1   least in my mind, that you want to be north of.  Thank you.
 2             MS. HANLEY:  This has been very interesting.  Just
 3   to confirm, we don't know if we're going to do a pilot yet.
 4             MR. HEALY:  I just like to think positively.
 5             MS. HANLEY:  We were hoping to get some information
 6   from the third panel on whether that is the appropriate
 7   mechanism to study this.
 8             I thought some interesting things came up in this
 9   discussion.  In particular, the notion that perhaps we should
10   not at the Commission focus simply on small and mid-cap
11   stocks, that perhaps large cap stocks could benefit with a
12   closer look at tick sizes there, perhaps even going in the
13   opposite direction.
14             I wanted to hear what the panel had to say about
15   that.
16             MR. CONCANNON:  I do think if we are going to
17   choose a pilot, and when I look at the liquidity challenges
18   that we have in our market, it's at the lower end of the
19   market.  We have let a lot of listed companies come in, so
20   there are standards, but they are fairly low historically. 
21   There are some companies that probably shouldn't be listed. 
22   There are some companies that are going to trade poorly no
23   matter what we do.
24             There are thousands of companies that have not
25   traded well for retail or for the buy side.  More
0129
 1   importantly, the buy side has had the biggest challenge
 2   trading these mid-cap and less liquid stocks.
 3             I do think when we think about our market
 4   structure, these stocks have been treated the same way our
 5   most liquid largest caps have been treated.
 6             When I think about who deserves the pilot and where
 7   we will get the best economic studies from, I do think
 8   focusing on a simple group of less liquid names as opposed to
 9   sticking in some large cap names that people aren't
10   complaining about, there is plenty of liquidity, we might be
11   putting those names in for a different reason, and we can
12   impact other things by putting those names in.
13             I think keeping the pilot simple, keeping it fairly
14   targeted will actually be a successful pilot and we will
15   actually get it done.  If we complicate it with all these
16   other market structure potential's, then we may not have a
17   pilot.
18             MS. HANLEY:  The pilot is one issue.  I think you
19   mentioned before that you thought there were some issues in
20   the larger cap, particularly even in tick sizes, that could
21   be addressed by the Commission.
22             I am interested to hear whether the one penny tick
23   is fine at the large end and we should just leave it there or
24   should we go to sub-pennies.  What do you think?  
25             Are there any issues there, not necessarily as part
0130
 1   of looking at small and mid-cap stocks, that may be a
 2   different examination, but since we have the opportunity here
 3   to talk about tick sizes in general, whether any focus should
 4   be paid to the larger end of the market.
 5             MR. SHILLMAN:  In connection with that, there has
 6   been a lot of talk about maker/taker and if we were to go to
 7   smaller tick sizes for the most liquid stocks and perhaps
 8   couple that with a re-evaluation of maker/taker, would that
 9   be something worthwhile doing.
10             MR. CONCANNON:  I guess the first question is what
11   are we trying to solve.  Is it a liquidity issue in the large
12   cap stocks or is it an economic issue.
13             I start from the premise that we have to look at
14   every stock individually or the characteristics of all our
15   stocks, both price, liquidity, float, and how they trade, and
16   determine the right market structure which includes trade
17   increment.
18             There is this organic thing called a "market," and
19   there are different instruments trading in it, and we have to
20   constantly fine tune that.  We haven't touched it in over ten
21   years.
22             I think looking at large cap, small cap, I agree,
23   we have to look at it.  I think the effective spread in some
24   of the large cap names suggest the market could deal with a
25   smaller tick size.
0131
 1             I do think we start bumping into the original
 2   issues that we dealt with in Reg NMS, which was does everyone
 3   have access to that lower tick size, and is it fair access,
 4   did they see it, can they trade it.
 5             I think with the sub-penny stocks, maybe we have
 6   found a way to do that.
 7             It does raise other issues around how to get to the
 8   lower tick size in the larger cap stocks.  
 9             I guess what I'm saying is a pilot on the small end
10   will reveal a lot of information whether there are benefits
11   to liquidity or not, and that may result in changing things
12   across all stocks, large and small.
13             MR. ISAACSON:  I think we would recommend from
14   BATS' perspective that we have probably a separate pilot for
15   very liquid or lower priced securities that appear to be tick
16   size constrained, but we would do it as part of this holistic
17   review of tick sizes that we haven't really looked at in over
18   a decade.
19             I refer back to the AMS study, this notion of
20   spread leeway, which really drove the discussion in Europe
21   when we talked about putting in their tick size regimes, what
22   is the optimal spread leeway or the number of ticks in
23   between the average quoted spread.  It's a very academic
24   thing to think about.
25             We think the number should be at least one.  You
0132
 1   can argue about whether it's one to two or one to five.  We
 2   think it ought to be at least one.  In most highly liquid
 3   securities in this country, it's zero, because it's a penny
 4   wide every day, and that is good for investors who are only
 5   paying a penny wide spread.
 6             Paul is offering them great executions and Kevin,
 7   through an institution, is getting the advantage of that as
 8   well.  Potentially, that is where most of the trading occurs
 9   as well.  
10             I think if you do a pilot that is separate from
11   trying to solve the small or mid-cap problem, you can
12   actually give decreased costs to investors even further with
13   the large amount of trading that's going on in the U.S.
14             MR. JIGANTI:  We are talking about three issues
15   here.  Within the liquidity portion of this, we are talking
16   about the low end, the lower micro cap end.  We are talking
17   about sub-pennies in the super liquid, let's call it the Bank
18   of America's of the world, and then we will talk about the
19   Apple/Google world.
20             Three completely different panels, three sets of
21   data.  Not to be confused with sprinkling any market
22   structure changes like maker/taker changes.
23             My concern is really the retail investor, of
24   course, what are they going to do when they look at the
25   screen and they see a half tick wide spread.  I don't know
0133
 1   that gets them to where they feel more confident in the
 2   marketplace.  Is it going to create a little more
 3   "flickering?"  Hopefully, not.  Presumably, in those super
 4   liquid's, you probably wouldn't have that.
 5             Then challenging the very high priced stocks when
 6   notionally, you are talking about a much smaller spread. 
 7   Apple being 40 times larger than Bank of America, presumably
 8   that is a 40 cents spread.
 9             I would argue it has completely different trading
10   characteristics and the retail customer has the ability
11   through a firm like TD Ameritrade use every one of the venues
12   that is out there to get the best execution that is
13   available.
14             MR. HATHEWAY:  As I have said a number of times, as
15   have other people, I don't think one size fits all works
16   particularly well.  The European experience shows we can
17   manage as an industry multiple tick sizes.  We can manage
18   maker/ticker fees with structures other than 30 cents per 100
19   to take.  We do that in the States below $1.
20             Reconsidering a lot of things that we put in place,
21   as Chris said, a decade ago, certainly makes sense.  
22             When you have a highly retail focused security, big
23   institutional following as well, 100,000 price changes a day,
24   100 shares on either side of the market, similarly situated
25   company, 80 price changes a day, 100,000 shares on each side
0134
 1   of the market, the stock price difference isn't a factor of
 2   1,000, it's more like a factor of 100.  It is essentially ten
 3   times excess price changes in high priced stocks and ten
 4   times cued up liquidity in the low priced stocks. 
 5             It just doesn't seem that a penny works for
 6   everything.  If re-evaluating pennies also means it's time to
 7   re-evaluate some of the other things that were put in place,
 8   in large part, in the name of simplicity, we were manual when
 9   we did decimals.  We were manual when we did NMS.
10             Poor Chris had to keep all those price schedules in
11   his head.  Now, it's all programmed.  You can sit there and
12   talk to panels like this and not worry about actually trying
13   to make money.
14             MR. CRONIN:  He's not poor old Chris any more. 
15   He's just old Chris now.
16             (Laughter.)
17             MR. CRONIN:  I think --
18             MR. CONCANNON:  Can we focus on decimalization,
19   please?
20             (Laughter.)
21             MR. CRONIN:  I've seen the numbers of the high
22   frequency guys.  Come on.
23             I think our thoughts on sub-penny pricing is at
24   best we are dubious, as you might guess.  With a move to sub-
25   pennies, and I can understand why the Exchanges would
0135
 1   propagate such a move because it will likely result in more
 2   activity, and at the end of the day, that's how they get paid
 3   -- this activity is entirely, at least in my mind, very, very
 4   difficult to determine whether or not it is valuable or not.
 5             At the end of the day, trades can just be trades. 
 6   It doesn't necessarily equate to liquidity.  I think that is
 7   what you get.
 8             From our perspective, giving somebody price
 9   improvement of a tenth of a cent on 100 shares is not
10   material.  Certainly, relative to the potential impact that
11   we can have in the market structure by allowing that kind of
12   behavior, so specifically this whole notion of crowding out.
13             We want to post a bid or an offer, so now we have
14   taken the minimum risk increment to do so from a penny to a
15   tenth of a cent, or essentially the increment that you can
16   make back by the rebate.
17             If anybody doesn't think that's going to ramp up
18   activity, you are crazy.
19             What does that mean?  You could certainly have so
20   much more quote volume that you really have issues with the
21   transmitability of data and by extension could get a lot more
22   of these technology driven problems that we have intra-day.
23             I suspect while it may be a good idea for some and
24   it might, at this juncture, it seems like an incredibly bad
25   idea to move the increment down to a sub-penny, certainly
0136
 1   from our perspective.
 2             As you described today, people are already getting
 3   the benefit of trading less than a penny.  Maybe that is in
 4   crossing networks, the trade at mid-point.  That seems like a
 5   reasonable outcome.  It still is only 50 cents on 100 shares
 6   traded.  Probably meaningful at some point if you do enough
 7   trades.
 8             MR. CONCANNON:  One thing I would add to that, and
 9   that is something that has not come up, trade increment does
10   impact order to trade ratio.  It does impact quote data.  
11             BATS has it in this study in Europe.  If you touch
12   trade increments of liquid stocks, you are playing with
13   market data, and the consumption of market data and fair
14   access to market data.  Touching trade increments, decreasing
15   trade increments, will increase order to trade ratio's, will
16   increase quote traffic.
17             We need to be mindful of that.  Touching trade
18   increments at the less liquid, you just don't have the number
19   of quote updates in less liquid names, so there is no
20   structural impact of that change in terms of a small and mid-
21   cap pilot.
22             MR. BURNS:  Can I throw a quick question in?
23             MR. HEALY:  If I could just add something before
24   that.  I just want to echo Kevin's comments here from a
25   little different perspective.
0137
 1             You heard my opening comments that one of the
 2   concerns of issuers is the markets will become a casino.  I
 3   will tell you the perspective of most issuers I speak with is
 4   if you go to sub-penny pricing, which very few of them
 5   understand, the casino just got larger.  I don't think they
 6   have any interest in the casino getting larger.
 7             MR. HATHEWAY:  It's sub-penny quoting essentially
 8   we are talking about.
 9             MR. HEALY:  It's the wrong direction.
10             MR. BURNS:  We got down the rabbit hole of the
11   pilot a little bit here, but recognizing we are going to lose
12   some of the panelists who have had some experiences in
13   Europe, I wonder if anyone can offer observations about
14   lessons learned, pitfalls, experiences.
15             You have multiple jurisdictions sometimes doing
16   slightly different things.  
17             Anything you would offer?
18             MR. HATHEWAY:  I think one of the key take away's
19   from us in Europe is the importance of uniformity.  That is
20   something I said in my other remarks we don't particularly
21   have here, even though we have the penny increment we have
22   been talking about.
23             The schedules worked.  We don't have a problem, we
24   are not aware of problems with member firms, and managing
25   schedules -- there are certainly behavioral changes that
0138
 1   occur when stocks pass a price increment that moves them from
 2   one tick to another.  Changes in the professional community,
 3   changes in the sort of investor community, which is one of
 4   the things the European regulators need to consider.
 5             Also, what was applied in Europe originally with
 6   the one exception of the U.K., was simply price based.  I
 7   think that was probably a little simplistic.
 8             Generally, we found the European experience to be
 9   very instructive for us in terms of a good learn about how
10   things could be done differently.  
11             It's probably not a model we could use for a study,
12   Maureen, because there are too many other differences.  I
13   think it is a very positive approach to dealing with the
14   price increment and trying to take some of the factors out of
15   the equation that are really not central to what it should
16   cost or what the appropriate cost of trading ought to be.
17             MR. ISAACSON:  I would just add to Frank's comments
18   that I think in the U.K., where there is some liquidity
19   component and not just price, I think that has been
20   instructive for us, and we would favor -- it is also to
21   decide what is a high liquidity security in the U.K., there
22   is a periodic evaluation of whether or not the security is
23   high liquidity.
24             I think making sure that evaluation period is long
25   enough is important.  You don't want securities going in and
0139
 1   out of a certain liquidity profile group.
 2             All and all, I think it has been a successful
 3   venture, and the uniformity is really paramount, you can't be
 4   competing on tick size.  I don't think as Exchanges that is
 5   really something we should be competing on.
 6             I want to make one other point.  While there is
 7   potentially some concern about sub-penny quoting, I
 8   appreciate Kevin's comments about is price improvement of a
 9   tenth of a penny really material, I would actually agree.
10             What was troubling to us is if you look at off
11   Exchange venues today, the amount of price improvement that
12   is actually occurring on those off Exchange venues is less
13   than that in most cases.
14             It begs the question of if you're going to change
15   quoting increment, you probably need to change trading
16   increment.  That is what we proposed for a pilot for small
17   and mid-cap for a nickel quoting increment that you would
18   have a trading increment of a full penny to force people to
19   not be offering de minimis amounts of price improvement,
20   immaterial amounts of price improvement.
21             I think that is really what is going on in our
22   market in the high liquid securities today where you have
23   people offering de minimis amounts of price improvement and
24   an artificially wide quote.
25             MR. CRONIN:  I think we would agree with the
0140
 1   concept that there is today a lot of internalization that
 2   takes place within the quote which is immaterial, so that
 3   price improvement that is being provided to orders that never
 4   see the light of day in the marketplace that are being price
 5   improved for again increments that I don't think many people
 6   would come to the conclusion is relevant.
 7             That gets to bigger internalization issues.  Again,
 8   I'm trying not to complicate any potential pilot program that
 9   would be done, but I do believe as I mentioned in my opening
10   statement that decimals, while an important concept here, is
11   not the only thing that probably should be looked at by
12   regulators.  Internalization is an issue.  
13             Could you get to a point where we agreed that if
14   there is price improvement on a security that takes place
15   mid-point, is that good enough?  Does it have to be an
16   increment better than that?
17             I think that would be a reasonable discussion to
18   have, but what has happened is these things continue to
19   happen in a vacuum with no aligned economic self interest of
20   everyone, so you don't get a common answer that would
21   probably be the one for the common good.
22             MR. ISAACSON:  I think that the $15 million that my
23   clients have achieved in price improvement back to their
24   pockets, back into the economy, is significant.
25             MR. GRAY:  I guess I would note we have already had
0141
 1   some really interesting and useful submissions in the comment
 2   file for quantitative ways to think about tick sizes and what
 3   they should be and perhaps how to do a pilot.
 4             It seems like there are two really differing
 5   approaches.  One is the spread leeway approach that has been
 6   discussed here.  Another one, which was probably focused more
 7   on at the earlier panel, is the clumping, a liquidity
 8   clumping approach.
 9             I would assume that any pilot would try to evaluate
10   those two ways of thinking about tick sizes.
11             My question for the panel is how would we know?  We
12   talked about liquidity or not.  If we had this pilot,
13   presuming we had it and you tested the clumping versus spread
14   leeway, what is the prime metrics you would look at to
15   determine whether liquidity got better or worse, and would
16   they differ depending on whether it was a large cap versus
17   small cap company?
18             MR. CRONIN:  I think defining success in this thing
19   is kind of an important concept.  I think unfortunately you
20   are likely to find that as you ask each one of us, we might
21   define that either a little bit different or radically
22   different.
23             If you start with the premise that institutions are
24   the most important, we would certainly acquiesce and tell you
25   that we would be okay with that.  I'm kidding.  Thanks for
0142
 1   laughing.
 2             (Laughter.)
 3             MR. CRONIN:  We have very good measures of what
 4   costs in trading are, as I hopefully gave some insight into
 5   earlier.  Spread isn't one of them.  When we look at our
 6   costs, as we kind of aggregate the different parts of
 7   transaction costs that we incur, the smallest part of it is
 8   commission.  
 9             Commissions might be seven/eight basis points, but
10   market impact and the information leakage that results from
11   large institutional trades entering the marketplace delay all
12   the different things that can happen, front running orders,
13   most certainly. 
14             Those kinds of transaction costs can accrue to
15   50/60/70/80/100/150 basis points in certain securities.
16             From our perspective, we have a reasonable baseline
17   to draw upon.  By the way, that is independent of volatility. 
18   I think one of the dirty secrets is people talk about
19   transaction costs being lowered dramatically, and they have
20   been in many respects over the last ten years.
21             Volatility is the number one determinant of
22   transaction costs for institutions.  As you lower it, and
23   let's be clear, over the last ten years, if you were to look
24   at nominal volatility, it was clearly lower now than then, we
25   would expect costs to go down.
0143
 1             Is that necessarily a market structure thing or is
 2   that something that was more related to just the volatility
 3   environment more broadly.
 4             Separating out volatility from the other parts of
 5   our transaction costs is actually kind of an important
 6   concept, but to my point, I think there are reasonable 
 7   metrics certainly that firms like Maureen's provide, ITG,
 8   that can give us very good insight into the efficacy of this
 9   effort.
10             As I said earlier, if it is that there is much less
11   price impact as we go in because the increment of risk goes
12   from say a penny to five cents, and as such, we have a lot
13   more incentive or maybe it's less disincentive to post our
14   bids and offers because if somebody wants to jump in front,
15   it's a more meaningful economic increment that they have to
16   do or risk they take to do so.
17             We might find that as a consequence, there are more
18   posted bids and offers of much higher quality and duration
19   and size.
20             Those are the variables that matter to us.  If you
21   have 100 up market that's a penny wide, and that's just
22   flickering back and forth, what is the value of that.  Is it
23   really an effective spread market of a penny?  
24             I submit to you that when you go to reach over and
25   take the offering and it disappears before you can do that,
0144
 1   and by the way, it disappears on all 13 Exchanges at the same
 2   time, that's not really something that we could all say is a
 3   great market structure.  That is not something that really we
 4   could say is indicative of transaction costs.
 5             From an institutional perspective, I think there
 6   are very specific things that could be brought to the
 7   discussion to help academicians and others who would be the
 8   third parties to help opine on whether or not this worked, or
 9   in fact, whether it did work for institutions.
10             The other thing is I would hope you would see
11   obvious things like there are more companies coming into the
12   market, there is much more liquidity.
13             If you look at the ownership and the list of people
14   who bought these companies as opposed to what it was
15   historically, you have some pretty good circumstantial
16   evidence as well.
17             MS. O'HARA:  Can I follow up?  I think again it
18   comes back to what is the goal of this thing.  I agree with
19   Kevin, the metrics for evaluating things like what are
20   transaction costs and trading costs.
21             I completely agree that you looking at quoted
22   spreads as only a minor issue here.  It has to do with the
23   effective spreads, the amount of effective spreads.
24             There are some widely agreed upon metrics that we
25   can look at.  However, I think the interesting thing about
0145
 1   this study may actually not be captured by any of that.  
 2             If you listened to the first panel, the first panel
 3   was convinced that the main thing this would do would bring
 4   lots of people back into putting orders into the book and
 5   institutions who wouldn't have done it before are now going
 6   to populate the book, and the books are going to be really
 7   big, and everyone is going to be great.
 8             That's really an interesting question.  Who is
 9   going to be putting orders into the book if you change this,
10   and you cannot ascertain that by simply looking at the prices
11   and the quotes.
12             You actually need order data.  Again, let me put my
13   academic hat on here.  In order to be able to kind of
14   understand the ecology of liquidity, you actually have to
15   understand sort of how long orders stay on the book, who put
16   them there, where did they come from, who is more likely to
17   be cancelling.
18             That sort of data is available.  I'll put in a
19   little plug here for the NYSE who has provided data for their
20   market, trying to sort that out.  Hopefully, Frank will feel
21   similarly, willing to do it.
22             I would encourage if you are going to do a study,
23   recognize that it's not like the old days where orders are on
24   the book, they are all the same.  Orders on the book are not
25   all the same.
0146
 1             It would be very, very helpful if you are going to
 2   gauge success to understand if the greater liquidity you got
 3   came from a broadening of your participation in the book or
 4   it came from an increased willingness of Chris and his market
 5   makers to be willing to step up and just throw more and more
 6   orders on.
 7             I'm not saying necessarily one is better than the
 8   other.  When you think about the goals that you have been
 9   hearing from a lot of divergent group of people, there are
10   people who would not view having more high frequency market
11   makers just providing more liquidity as the desired goal.
12             I just raise that.  I think you need more data, and
13   I think you will need a much, much deeper analysis than just 
14   effective spreads.
15             MR. BURNS:  Professor O'Hara, you have given us a
16   great note to end on.  We can continue talking.  I hope
17   people will over lunch.  We will be back at 2:30 this
18   afternoon when we will talk a little bit more about a
19   potential pilot and what that might look like.
20             Thank you very much to this panel.  I am very loath
21   to cut us off here.  This has been very helpful.  We look
22   forward to continuing these conversations with you.
23             (Whereupon, at 1:02 p.m., a luncheon recess was
24   taken.)
25           A F T E R N O O N  S E S S I O N  (2:33 p.m.)
0147
 1   PANEL 3:  STUDYING THE EFFECTS OF ALTERNATIVE TICK SIZES
 2             MR. BURNS:  I think we're going to try and start
 3   the third panel.  Oh, we're missing the third panel.  No,
 4   here we are.  It's hard to see down this row here.
 5             After a terrific morning, two very lively,
 6   thoughtful panels, we are turning to the afternoon session,
 7   and where we study the effects of the alternative tick sizes,
 8   and for that I'm going to turn the floor over to our Chief
 9   Economist and the Director of RiskFin, as we affectionately
10   call his division, Craig Lewis.
11             MR. LEWIS:  Thank you, Jim.
12             I thought what I would do, what we're going to try
13   to do today is try to solicit views on what we are trying to
14   accomplish with the development of a possible pilot.  I
15   thought I would begin my remarks by discussing what I think a
16   good pilot program involves, and after that I will then turn
17   it over to the panelists.  Many of you have actually
18   introduced yourselves already, but I'll give you a chance to
19   re-introduce yourselves to those who are just tuning into the
20   Webcast now, and I will then begin by a lead-off question,
21   and we'll go from there.
22             So the way I wanted to think about pilots is one of
23   the things that we do in the Division of Risk Strategy and
24   Financial Innovation is we help work and develop the economic
25   analysis that supports policy decisions and rule writing. 
0148
 1   Part of good economic analysis has four elements.
 2             The first element is to identify a market failure,
 3   understand what it is, what the problem is that you're trying
 4   to solve.
 5             The next is once you have identified that market
 6   failure, your next step is to identify a baseline from which
 7   you can begin to design a regulatory solution or, in this
 8   case, to design an appropriately scoped pilot program.
 9             Having identified the baseline, one can then
10   discuss what the economic effects a pilot program is designed
11   to address a market failure so we can better understand what
12   the market failure is, but we are interested in knowing what
13   the benefits, potential benefits, from this are as well as
14   the costs to market participants, as well as those
15   registrants that are actually required to comply with a pilot
16   program.
17             And then finally, the last piece of a good economic
18   analysis is to consider the alternatives and what the actual
19   economic implications are of the choice that you're
20   recommending or the study that you're conducting, as well as
21   what other possible policy choices could be made.
22             So this morning, we heard a lot of discussion about
23   what the problem is, and I felt that coming away from this
24   morning's two panels, there were a number of issues that
25   appeared to be on the table.  The first session focused on
0149
 1   the role of capital formation, the public market,
 2   particularly that formations of equity, the role of research,
 3   equity research, and how market structure issues might impact
 4   the provision of equity research and how that might
 5   ultimately affect the security issue process.
 6             The second panel we had this morning focused more
 7   on what I felt were market microstructure issues:  liquidity,
 8   the provision of liquidity, how more liquid markets might be
 9   able to support additional components like equity research.
10             So when you are, from my perspective, trying to
11   design a pilot, I'm looking for feedback and input on your
12   ideas of what might be the effective way to put one of these
13   together.  We want to understand what exactly it is that
14   we're trying to accomplish with a pilot program.  So that
15   would be the first objective of having a pilot study.
16             So understand what we're trying to solve and then
17   some issues came up.  Once we have a clearly articulated
18   scope for the project, the next came down to more
19   methodological issues, sort of econometric issues related to
20   how we would go about constructing a pilot that could
21   actually answer the questions we have in mind.
22             And so we had some concerns about how would you
23   identify a treatment group; what would be the right control
24   group; what are the factors one would consider when designing
25   an appropriate control.
0150
 1             Maureen raised a point about voluntary
 2   participation in a pilot program and some of the problems
 3   associated with allowing a pilot to be constructed by looking
 4   for volunteers versus a more scientific approach would be to
 5   go out and actually not only identify the control group, but
 6   to identify the treatment group as well.
 7             So there are a lot of issues on the table that
 8   we're interested in hearing your feedback about, and then it
 9   comes down to once you have these samples, keep in mind that
10   one of the things I think is important when you are designing
11   and trying to construct a control group, is that the control
12   group should be calibrated to actually answering the
13   questions that are on the table.
14             So if the questions that you're seeking to address
15   in a pilot program are around liquidity, the control group
16   should be focused around liquidity.  If it's designed to be
17   addressing equity issues, it should be looking at equity
18   research potentially as one of the potential controls.
19             So with that in mind, I think I'd just like to turn
20   it over and let everyone introduce themselves.
21             Maureen.
22             MS. O'HARA:  Well, thank you.  I'm Maureen O'Hara. 
23   I'm the Robert Purcell Professor of Finance at Cornell.  I've
24   spent a large part of my career studying issues in market
25   microstructure, and I've worked on a wide range of topics,
0151
 1   including the effects of fragmentation on market quality, the
 2   role of different market mechanism in impounding information,
 3   the nature of trading platforms.  Most recently I've been
 4   working on issues relating to high frequency trading. 
 5             I also am Chairman of the Board of ITG, which is a
 6   global agency brokerage firm.
 7             MR. JIGANTI:  Good afternoon.  I thank you for
 8   having me.
 9             I am Paul Jiganti.  I'm Managing Director at TD
10   Ameritrade in charge of micro market structure and customer
11   advocacy.  I have been in the trading business for 25 years
12   before coming over to TD Ameritrade and working in my current
13   role.  Most of my time, 16 years was with Susquehanna and a
14   year at NASDAQ.
15             I look at this potential pilot program as a great
16   data source and the ability to go forward and make some real
17   judgments of how the market could look going forward.
18             MR. WOMACK:  My name is Kent Womack.  I am the
19   Manulife Chair in Financial Services at the University of
20   Toronto, after 16 years at the Tuck School at Dartmouth.  I
21   was previously at Goldman Sachs where I did a lot of selling
22   of the IPOs and small companies that we're talking about
23   here.
24             My research has been in the fields of IPOs and
25   investment research, specifically, the value of analyst
0152
 1   recommendations, and so I will likely talk about that in a
 2   little bit.
 3             MR. REED:  I am Adam Reed.  I'm the Julian Price
 4   Scholar and Associate Professor of Finance at UNC
 5   Kenan-Flagler Business School.  I have a Ph.D. in finance
 6   from Wharton, and I've worked at the Federal Reserve for a
 7   couple of years before the Ph.D.
 8             I've studied a large range of trading issues,
 9   including issues on mutual funds.  I've had a special focus
10   on short selling and equity lending, and in that capacity
11   I've looked at a number of regulatory changes and the effects
12   of those changes on short selling, and benefitted from the
13   Commission's SHO pilot on short selling.
14             I've also done a lot of work on IPOs from both the
15   perspective of short selling and also from the perspective of
16   innovations that go into IPOs.
17             MR. SACHS:  Hi.  I'm Steve Sachs.  I'm head of
18   capital markets for ProShare Advisors.  I joined ProShares
19   about 18 months ago after a career in equity and derivatives
20   trading that spanned institutional buy side firms from
21   fundamental stock pickers to quantitative shops, you know,
22   small, mid, large cap across, you know, really the sort of
23   full size spectrum of an institutional money manager.
24             And I certainly want to thank the agency for
25   holding this roundtable today on obviously an extremely
0153
 1   important topic, not just this individual topic, but a lot of
 2   the things that came out in the first couple of panel, in
 3   particular, about just the broad sort of state of the market
 4   structure in this country and certainly taking a look at this
 5   particular issue is very timely.
 6             MR. HATHEWAY:  Thank you.  I'm Frank Hatheway and
 7   the Chief Economist of NASDAQ OMX.  It's a role I've had for
 8   about the last 12 years.  Prior to that I was a finance
 9   professor at Penn State University, and prior to that I was a
10   floor trader in the options industry at the Philadelphia
11   Stock Exchange.
12             I had a long set of experiences with pilots of
13   various types.  A short sale pilot which Adam referred to
14   under Reg. SHO was, of course, one.  Penny pilots in options;
15   my group will be performing a great deal of analysis on the
16   upcoming limit up/limit down pilot as one of the follow-on
17   initiatives of the Commission in response to the flash crash
18   of May 6th.
19             I think pilots are a terrific way for the
20   Commission and market participants to gather information
21   about alternative mechanisms that might be used in the U.S.
22   securities markets, and I look forward to a more detailed
23   discussion of the proposal under consideration here.
24             Thank you.
25             MS. McCARTHY:  I'm Maureen McCarthy.  I'm the
0154
 1   Director of Sales and Trading at JMP Securities.  JMP
 2   Securities is a San Francisco headquartered, equity focused
 3   investment bank, focused on research.  We have 350 names
 4   under coverage approximately, and 87 percent of those are in
 5   the micro to mid-capitalization.
 6             We worked on 64 follow-ons and eight IPOs in 2012,
 7   and have raised 75 billion for our clients at an average deal
 8   size of 142 million.
 9             Most of my career has been in the mid to micro cap
10   space.  I have made markets for Kidder Peabody in the late
11   '80s and then for Montgomery Securities in the mid-'90s.  I
12   was the head of trading and eventually sales trading for
13   Roberts & Stevens from 1996 to 2002, and I've been at JMP
14   Securities for nine years.
15             And, incidentally, JMP Securities is itself a micro
16   cap publicly traded New York Stock Exchange stock that is
17   majority held by JMP Group, JMP.
18             MR. CLARK:  Good afternoon.  My name is Collin
19   Clark, and I am Senior Vice President of the Strategic
20   Analysis and Market Data at NYSE Euronext, where I lead the
21   data analysis team in the areas such as market quality,
22   client behavior and pricing.
23             I think I'd like to, you know, first of all
24   congratulate Frank for joining the company. 
25             (Laughter.)
0155
 1             MR. CLARK:  I was not aware of that.  
 2             I've been at the exchange for seven years.  Prior
 3   to that, I was an equity research analyst at Merrill Lynch
 4   and Citigroup for six years covering brokers, asset managers,
 5   exchanges, trading companies.  Prior to that I worked at
 6   Fidelity for five years where I had various roles, including
 7   worked as a mutual fund analyst.
 8             I'd just like to spend a couple minutes just making
 9   the panel aware that NYSE Euronext provided a written
10   statement in support for the pilot that increases tick sizes
11   in less liquid securities, and we believe this could have a
12   positive effect on depth and liquidity.  From an issuer and
13   investor perspective, there is a lack of liquidity in
14   thousands of stocks in our markets.  This leads to more
15   volatility and makes it more difficult to attract
16   institutional investors and analyst coverage.
17             From a market maker perspective, the current market
18   structure is not properly incentivizing brokers to display
19   liquidity, and we believe changing the tick size could help
20   to consolidate liquidity at wider price points thus reducing
21   undercutting strategies.
22             I want to thank the Commission for inviting me to
23   participate in today's panel and the staff for organizing a
24   review of the topic that we believe is deserving of
25   attention.
0156
 1             Thank you.
 2             MR. LEWIS:  Okay.  I'd like to thank all the
 3   panelists for agreeing to come down here and spend some time
 4   with us today.  
 5             I thought before we move on since we're coming back
 6   after lunch, I will also introduce the rest of my colleagues
 7   that are from the SEC staff.  To my immediate left or your
 8   far right would be Lona Nallengara, who is the Director of
 9   the Division of Corporation Finance here at the Commission.
10             To my right is Amy Edwards, who is an Assistant
11   Director in Division of Risk Strategy and Financial
12   Innovation, and she is responsible for the Office of Markets.
13             We have Kathleen Hanley, who is my Deputy Director
14   in RiskFin, as well as Deputy Chief Economist.
15             At the far end of the table we have Gregg Berman,
16   who is leading the Office of Analytics and Research.
17             Oh, Jim, you're hiding back there.  My apologies. 
18   That was the visual contact I was looking for.
19             So to my far -- is Jim Burns, Deputy Director of
20   Trading and Markets; David Shillman, Heather Seidel, and Dan
21   Gray.
22             So thank everybody and look forward to your
23   comments.
24             So I guess I will start out by throwing the first
25   question over the transom, and what we're doing is
0157
 1   considering the first two panels.  What information would
 2   help us better understand the effect of tick sizes on all or
 3   a subset of stocks?  What alternatives to a pilot should the
 4   Commission consider in gathering additional information on
 5   the effect of tick sizes?
 6             One of the things I'd be interested in hearing from
 7   academic colleagues would be do you feel that there are any
 8   gaps in the existing academic research that the pilot could
 9   be used to fill.
10             MS. O'HARA:  I'll start us off.
11             There are huge gaps in what we know in academia
12   about markets because so many of the studies that we have
13   done in the past that have been referenced by you and others
14   are necessarily dated, and particularly a lot of the studies
15   that were done about decimalization.  There's a lot of
16   excellent studies, but they were done, many of them, in, you
17   know, the early 2000s, which is really in the period before
18   the high frequency world began.
19             So academics love pilots, right?  Because we would
20   love to know more about what's going on in markets, and if
21   you would design a careful experiment that we could analyze,
22   we're always happy to do that.
23             There is, however, sort of an interesting set of
24   questions in the sense that I don't know that we actually
25   know that much about the nature of the liquidity provision
0158
 1   process and markets right now as it is, and you know, the
 2   ability to do research these days in markets and to
 3   understand what you're looking at is non-trivial because it
 4   requires data that most academics just cannot get.  Tack data
 5   on a monthly basis is not sufficient to address these issues,
 6   and so we've been very fortunate in academia when the markets
 7   have been willing to support this.
 8             And, you know, Frank Hatheway and NASDAQ provided
 9   an HF database which is very handy.  The New York Stock
10   Exchange has been helping me with some work on trying to
11   understand the nature of liquidity at the NYSE.  So I think a
12   pilot can be very, very useful, but it may also be helpful to
13   be able to have data available that allows people to just
14   evaluate the current structures of market because within
15   those markets, for example, the tick size actually, the
16   relative tick size, does differ across stocks.  If you can
17   match stocks that have the same characteristic but one is a
18   $4 stock and one is a $40 stock, then you actually have a
19   situation where there's a relative tick size that's ten times
20   as large as the other one.
21             So should you decide not to do a pilot, I would
22   like to hope that maybe you could encourage organizations to
23   step up as the NYSE and NASDAQ have done to try and provide
24   data that could help, and obviously if you do do a pilot,
25   then the issue is, you know, to be very, very careful how
0159
 1   it's structured so that you don't get caught up in the
 2   plethora of issues that really surround liquidity provision
 3   in markets now.
 4             MR. LEWIS:  Adam.
 5             MR. REED:  Yes.  I would say that a pilot is
 6   probably one of the best, if not the best, ways to study this
 7   issue.  I think in my mind the key defining feature of a
 8   pilot is that there's a built in randomization that you
 9   basically can't get except for the luckiest of natural
10   experiments, and I think, you know, the Commission's
11   willingness to design a pilot, I think, definitely goes in
12   the right direction.
13             From my experience with the Reg. SHO pilot, I think
14   it provided a good test bench for lots of issues around short
15   selling, and I would say it even went past that.  The people
16   sort of think of it as a nice instrument to use in
17   identifying liquidity overall, and they'll think about some
18   of the work that came out of that pilot study as a way of
19   testing various hypotheses even beyond the direct hypotheses
20   about short selling, in particular.
21             MR. CLARK:  If I may just highlight, actually, an
22   academic paper done during -- it was a 1999 tick change that
23   happened on the Paris Bourse where some securities had lower
24   ticks and some had higher ticks, so kind of a unique event,
25   and the conclusions were interesting, and that might be a
0160
 1   worthwhile review if you haven't read it, but you know, the
 2   summary, at least as far as stocks that had higher tick
 3   increases, the outcome was limited to no impact on spreads
 4   and increased displayed liquidity in that market, so just one
 5   study to note.
 6             MR. LEWIS:  Let me ask a related question.  If you
 7   were to identify the market failure that a pilot study is
 8   going to be designed to answer, in your view, what is that
 9   market failure?  What is the pilot designed to look at?
10             MR. SACHS:  I can certainly jump in from the
11   practitioner's perspective.  
12             I think that there's actually really multiple
13   things, but if I had to prioritize them and rank them,
14   displayed liquidity, from the institutional investor's
15   perspective and the access -- in the ability to access that
16   displayed liquidity is probably at the top of the list.
17             One of the things that, you know, we see already in
18   the data, you know, from the academic studies, and if you,
19   you know, talk to practitioners and traders of, particularly,
20   small and mid-cap securities, is that -- and it was alluded
21   to in one of the -- in the first panel this morning -- while
22   the tick size is a penny, the actual, you know, spread that
23   you're trading at in most of these securities is typically
24   far greater than a penny; it's something in the, you know,
25   probably five-to-eight-cent range, if I had to, you know,
0161
 1   anecdotally take a guess.
 2             So, the market, in and of itself, is already
 3   aggregating liquidity, you know, at the points where supply
 4   and demand, you know, meet
 5             One of the things that, you know, the institutional
 6   community keeps, you know, falling back on with the issues
 7   is, again, displayed liquidity, access to the liquidity or
 8   the -- the flickering issue, things of that nature.
 9             So, from a pilot program perspective, which I do
10   support, and I agree wholeheartedly that I really don't think
11   that there is a better way to gather actual data that we then
12   could use for a qualitative analysis, an empirical analysis,
13   that doing a pilot program.
14             There are multiple models around the globe that you
15   could use from a multi-tier, a multi-tick tiered system, but
16   ultimately, looking at a number of issues, but I would argue
17   the displayed liquidity issue and the -- the aggregation of
18   that displayed liquidity issue is certainly going to be first
19   and foremost from an institutional investor's perspective.
20             MR. HATHEWAY:  If I were -- I think I'm going to
21   rank some possibilities for market failure for things that
22   may be harder to investigate to easier.
23             In the panels earlier today, part of the issue, and
24   I think a key part of the issue that brings us here is the
25   change in the capital formation process for small companies.
0162
 1   Essentially that comes down to cost and availability of
 2   capital and public equity markets cease to be as attractive
 3   to small companies as it once was.
 4             Essentially, there is an increase in the
 5   illiquidity discount that small companies face.  So, the
 6   first market failure is an illiquidity discount inadvertently
 7   imposed by decimalization.
 8             I think that's very hard to observe and very hard
 9   to measure.
10             So, we're then left looking at proxies, of which
11   liquidity, as Steve was just saying -- and now I'm not
12   talking about an illiquidity discount that investors have to
13   take into account that raises the cost of capital of firms
14   but simply liquidity getting a trade done in the market is a
15   proxy for what might have happened to illiquidity discount. 
16   That can be investigated through a pilot.
17             The other potential market failure, as we were,
18   again, discussing this morning, is the coordination failure,
19   lack of uniformity around trading increments, and what impact
20   that may well have on incentives to display liquidity, on the
21   ability to access liquidity; heard very different points of
22   view this morning from different constituencies as to whether
23   the ability to trade on one set of increments in one type of
24   markets and a different set of increments in another type of
25   markets is beneficial or harmful or not.
0163
 1             The final failure is whether there is insufficient
 2   revenue associated with bringing small companies to market or
 3   having small companies represented as part of your sales
 4   portfolio for your sales force to fund research.
 5             As, again, some of the points made this morning I
 6   agree with, that's probably the most -- the least directly
 7   related to tick size and the most difficulty to explore with
 8   a pilot, cause there may well be a great many factors at
 9   work.
10             That's my list.
11             MR. BERMAN:  Could I ask a follow-up question?
12             Steve, you mentioned displayed liquidity.  If
13   there's 10,000 shares out to 10 cents today and then we
14   engage in the pilot, could you comment on the three different
15   aspects I think that -- that we could see in change, though.
16             One is the 10,000 stays.  Nothing changes.  The
17   second is that the 10,000 becomes 20,000, so it increases,
18   but the 20,000, I think you mentioned flickering or it's less
19   accessible.  So, how would we actually measure whether or not
20   that 20,000 is valuable compared to the original 10?
21             Or the 10,000 over 10 ticks becomes 5,000 at one
22   tick and then 5,000 at the next tick, so it's the same 10,000
23   but it's now aggregated, cause I think you mentioned
24   aggregation.
25             Do you think about each of those three differently,
0164
 1   and should there be different criteria?
 2             MR. SACHS:  I don't think that there should be
 3   different criteria, because what I'm discussing there is
 4   really more the reality, as you know, of how markets work and
 5   how market makers would think and, ultimately, how we've, you
 6   know, these days, programmed algorithms to -- to work and to
 7   think in the marketplace.
 8             But the assumption is, in the concept -- and maybe
 9   it's just nothing more than an assumption -- is that, you
10   know, again, given the -- the current one-penny tick
11   increment, there is an extremely low hurdle for a market
12   maker or other market participants to step into the process.
13             So, that, by its very nature, is going to encourage
14   people to step into the process, but again, given the -- the
15   market structure that we have and ultimately what we've moved
16   to post-Reg. NMS is that, you know, you don't really display
17   that liquidity, because why would you?
18             You can step into the middle of the quote for a
19   one-penny increment for as little as 100 shares or less,
20   which is exactly what your algorithm or market making, you
21   know, intention or tendency is.
22             So, if you -- if you think back to when there was a
23   -- you know, a 16th or 8th of a point spread, when the
24   economic incentive to step into the middle of that quote was
25   a much higher hurdle, the tendency of the market participants
0165
 1   was to display liquidity from a priority perspective.
 2             Because the incentive was not as great, the
 3   incentive to hide what you were doing in the marketplace, at
 4   that point, from a practitioner's perspective, was less.
 5             I do fully, fully agree or acknowledge that,
 6   ultimately, given the psychological change that's taking
 7   place with market participants today, that you may not see
 8   that net outcome, but ultimately I think it's one of the very
 9   important facts that a pilot program would prove out.
10             One of the baseline assumptions from the trading
11   community, institutional trading community, is that a wider
12   spread with lower -- with higher economic incentive to step
13   in the middle would create more displayed liquidity, you
14   know, from that perspective.
15             So, I don't think of those things differently.  I
16   would actually argue for keeping the variables around that
17   type of activity as few as possible to see what the possible
18   outcomes are.
19             MR. CLARK:  If I could just add, I think another
20   important point to make is that, you know, a higher tick does
21   not necessarily mean a wider spread, and that obviously we're
22   talking about this tradeoff between how much additional
23   displayed liquidity we can get as well as that -- that
24   tradeoff of the spread.
25             I mean, to give you an example -- and it's more of
0166
 1   a question -- you know, if you've got a stock that's got a
 2   quoted spread of, say, seven and you move the tick increment
 3   to five, does the -- does the stock actually start moving to
 4   the five increment and actually narrow the spread where you
 5   get some liquidity now aggregated at five and you get a lot
 6   of additional liquidity aggregated at 10, the net result
 7   being you've got a narrower spread and potentially more
 8   liquidity, but I think there's this -- this tradeoff between,
 9   obviously, spread and displayed liquidity that's really at
10   the core of this.
11             MR. HATHEWAY:  If I could make one point, I mean
12   one of the temptations of a question like that is to say you
13   could answer it with institutional trading cost analysis,
14   such as what Maureen's -- the company Maureen's chairman of
15   does in there full-time.
16             The trouble with that approach is it assumes that
17   there is no change in the mix of what type of stocks
18   institutions invest in.  It's typically transaction cost
19   analysis when you get summary statistics.  It's
20   volume-weighted.
21             The fact that there is a transaction has to have
22   occurred before you can measure transactions cost.  
23             If a pilot makes it less attractive to do a
24   transaction in certain types of costs, those stocks drop out
25   of the study.  If it makes it more attractive to do
0167
 1   transactions in stocks where there are not transactions
 2   today, they come into the study.
 3             It makes the analysis more complex.  Folks like
 4   Adam and Kent and Maureen certainly have the hardware to deal
 5   with it, but it becomes now more on the -- I won't say
 6   necessarily witchcraft but a more complex part of statistical
 7   and economic analysis to address what sounds like a simple
 8   question:  Did costs go up?  Yes/no.
 9             If the program actually works, you're going to get
10   more liquidity, more costs to measure, and stocks that people
11   don't trade today, cause they're too costly, and that may
12   pull averages up, but in effect, the average is masking,
13   really, what we're interested in.
14             MS. O'HARA:  Can I just echo a little bit what
15   Frank has brought on?  I think in any study like this, there
16   are going to be winners and losers in terms of the effects on
17   stocks, right?  
18             For the stock that had the seven-cent spread, there
19   may be a big winner in the sense that perhaps spreads go to
20   five.  The stock that had the three-cent spread -- its spread
21   might also go to five.
22             And so, one of the challenges in this is that any
23   -- any pilot has to be able to differentiate kind of what's
24   happening both to the winners and the losers, so averages can
25   be very misleading.
0168
 1             The second issue that is one to contemplate is, as
 2   we look at orders that trade in markets, we already know that
 3   an awful lot of liquidity now doesn't trade on the exchanges.
 4   It trades elsewhere in various other settings.
 5             If you widen the spread, the tick, right, then for
 6   many stocks, any stock right now that's trading less than
 7   that, the spread is going to be bigger, and that gives
 8   incentives for things to trade off exchange.
 9             So, if we define the metric just simply have we
10   lowered trading costs, we can answer that.  If you define the
11   metric, have we changed the liquidity provision process
12   whereby now people who were not willing to post limit orders
13   before but are now, so books are bigger, then you need a lot
14   more data than just looking at the trades and the quote.
15             You really do need to look more carefully at the
16   nature of the liquidity on the book, and you can do that if
17   you have the data.
18             But I would suggest if you're going to do this
19   study that you really shouldn't just content yourself with
20   trying just to measure the transactions costs without using
21   the opportunity to understand more fully the liquidity
22   production process.
23             MS. EDWARDS:  I think that's a good transition to
24   the next part of this panel, where we talk about a bit of
25   what the goals of a pilot should be.  What do you recommend
0169
 1   the goals of a pilot should be if the Commission considers
 2   going in that direction?
 3             And we'll transition into design from the goals,
 4   but at least for right now, let's think about what goals
 5   should a pilot be designed around?
 6             MR. JIGANTI:  I'll jump off and say do no harm. 
 7   You know, in a lot of these stocks, even in this end of the
 8   curve, the typical retail customer gets -- 96 percent of the
 9   time they get filled on their full order at the quote they
10   see.
11             So, you know, it's hard to give up the extra spread
12   for no apparent reason, for, you know, the group that I am
13   here for.
14             So, you know, do no harm.
15             MR. SACHS:  Obviously the balance to that -- Kevin
16   Cronin on one of the earlier panels stated it very well. 
17   Again, the vast majority of individual investors in this
18   country get their investment exposure through some sort of
19   pooled vehicle or some sort of professional managed vehicle.
20             So, from a design perspective, obviously, you know,
21   being extremely biased and close to this issue, I would
22   certainly want to see a very large dose of balance amongst
23   the constituents of what the ultimate, you know, effects
24   would be.
25             I certainly echo the effect of do no harm, but
0170
 1   ultimately I think that is, one, somewhat unrealistic, but
 2   two, you know, at the end of the day, you know, most
 3   regulatory actions and most market structure actions have
 4   some sort of reaction.  Some are positive; some are negative.
 5             So, ultimately, from a design perspective -- and to
 6   my point, a heavy dose of balance for all of the constituents
 7   and looking at it from a much bigger picture perspective, as
 8   opposed to just an individual transaction or a quote.
 9             One of the things that I wrote down earlier that
10   came up in an earlier panel is, all market participants,
11   myself among them, we tend to confuse quote with liquidity on
12   a regular basis, and I caution us, in the design of a pilot
13   program, to not be so narrowly focused that we make that --
14   we make that confusion.
15             MR. GRAY:  I think that's a good point, and I guess
16   I would ask Maureen, based her everyday experience with
17   trading small cap stocks, if there was a pilot that didn't
18   widen the spreads for a number of the -- widen the increment
19   for those stocks, what are the sorts of things that you would
20   focus on to determine whether the market was better and
21   liquidity improved in those stocks, beyond perhaps just
22   looking at the quoted spread?  Are there other metrics that
23   you would recommend?
24             MS. McCARTHY:  I think average -- average trade
25   size and not -- not -- what we would recommend, actually, is
0171
 1   -- I think somebody mentioned this in another panel, but --
 2   is having a -- a study group that is the pilot, having a --
 3   almost a paired group that is the non-pilot but a study group
 4   to compare it against, and then, also, of course, to compare
 5   it to the statistics of before -- before the pilot.
 6             In terms of -- in that regard, average trade size,
 7   dollar volume traded, volume traded, and then we thought of,
 8   you know, some other ways that might be outside of the box to
 9   figure out whether not only market makers are accepting their
10   obligations but also if it is actually working in terms of
11   adding liquidity, and some of that can't be measured.
12             If a market maker wants to add liquidity to the
13   market, we will send out an IOI to our institutional
14   customers, and that IOI can be acted upon by the institution,
15   or it may not be acted upon by the institution, but if it
16   isn't acted upon by the institution but there is more of a
17   willingness to trade than there was before because there is a
18   perceived wider spread and a better economic incentive, that
19   might not be measured in just the stats that we were talking
20   about earlier, average trade size, and it may.
21             But one way or one additional measure that's just a
22   little out of the box is we could look at IOIs in the control
23   group prior, IOIs now, and then compared to the IOIs in the
24   -- in the study group, as well, and whether -- hopefully that
25   data is available.
0172
 1             I think it is.  I think it's different from
 2   advertised trade volume by market makers, but -- but it is,
 3   in and of itself, a metric that we could use to determine
 4   whether people were more willing to -- and market makers were
 5   more willing to commit capital.
 6             MR. HATHEWAY:  I think one of the goals we've had
 7   here is capital raising for small companies, either ones that
 8   would like to do an IPO or ones that are currently listed and
 9   looking to raise capital through a follow on or PIPE or what
10   have you.
11             For a pilot of sort of reasonable length of time,
12   if we look at the really excellent work Grant Thornton has
13   done, it's clear that these changes are somewhat slow,
14   particularly with respect to, say, the short sale pilot or
15   the option penny pilot, where the impacts were relatively
16   quick.
17             So, it's not clear that we would have enough data
18   to answer questions like that at the end of a one-or-two-year
19   pilot.
20             A proxy that occurred to me while Maureen was
21   speaking is you can look at things like ownership mix, you
22   know, who is in these stocks that are existing in the market
23   today, who holds them, what's the institutional holding, how
24   diversified the institutional holdings are, things like that
25   that, again, are proxies for accessibility for the
0173
 1   institutional community, for the retail community into some
 2   of these securities.  That's all data that's available.
 3             MR. WOMACK:  You know, when decimalization came to
 4   be, it was about saving investors' bucks, right, big bucks, I
 5   would say, and you know, I think one of the questions we
 6   should have firmly in the front of our mind going forward on
 7   something like this is what are the transaction costs that we
 8   expect to be imposed on investors, if they are such, in order
 9   to find out, over how many years -- three, four, five -- that
10   there's no benefit?
11             I mean, I think a healthy question is, what is the
12   likelihood, the probability that there is, in fact, some
13   benefit to tinkering with the cost of transacting?
14             I don't think -- I'm not at all convinced that I
15   have a good guess about that.  I would guess that this is a
16   important question maybe this panel could opine on as to the
17   likelihood that after you've spent how many millions of
18   dollars on higher transaction costs, what do you think you're
19   getting out of it?
20             MR. HATHEWAY:  The panel hasn't started arguing
21   much yet, so maybe it will start.
22             When decimalization came in, spreads went up on a
23   quarter of our stocks, mostly inactive stocks, and so, this
24   is essentially Colin's point.  
25             It's not clear spreads, which is the retail cost,
0174
 1   generally speaking, are going to go up or down, and then the
 2   institutional costs are going to go up or down, and the nice
 3   thing about a pilot -- and the short sale pilot and the
 4   implementation process for decimals have this feature to
 5   them, as well -- you can pull the plug.
 6             If it really -- say we -- we do this pilot and DMMs
 7   all quit and the market makers all quit and these stocks are
 8   now just flat out not able to trade, we could change our
 9   minds about the pilot, but it's not -- again, to answer your
10   rhetorical question, why do we think things may get better --
11   well, because for the subset of stocks, things got worse, at
12   least in the NASDAQ world, very different from today, when we
13   did this -- we did a reverse 12 years ago.
14             So, I think there's some basis to think it may get
15   better.
16             MR. JIGANTI:  And in your mind, is that directly
17   related to decimalization, or could it have been other
18   economic forces that have been part of that explanation, too,
19   over the last 5 to 10 years?
20             MR. HATHEWAY:  Well, the change was very short-term
21   around the introduction of decimalization.  So, this was the
22   pre-post data analyses.
23             I think the -- probably the biggest weakness with
24   my statistic is I'll bet for inactive stocks there's a ton of
25   noise in what the typical spread is.
0175
 1             MS. O'HARA:  I would just add that if, in fact, the
 2   goal is to improve the liquidity for a subset of stocks, then
 3   clearly what you want to be looking at are the variables that
 4   would reflect that.
 5             So, for example, time-rated orders on the book,
 6   right?  If -- if you really think that part of the problem is
 7   that liquidity has been flickering, then -- and this is going
 8   to incent people to put orders in the book, we should be able
 9   to see that in the duration of orders on the book.
10             We should be able to see, for example, if the
11   argument is that institutions will now be incentivized to put
12   orders in the book, then when we look at -- right? -- who is
13   putting the orders in the book, we should be able to see more
14   institutions.
15             If, instead, what you're seeing is market makers
16   are now incentivized to put more orders on the book, that may
17   well lead to enhanced liquidity, but the people this morning
18   who really are hoping that these incentivized institutions
19   are going to take the extra returns they're making and use
20   that to move forward in research, if they're not actually the
21   source of any action here, that's unlikely to happen.
22             So, I think there's a variety of measures that we
23   know how to look at.  We certainly know how to measure
24   transactions cost.  We know how to measure relative expected
25   spreads.  
0176
 1             A lot of these actually turn out to be highly
 2   correlated with the dollar trading volume in a stock.  So,
 3   you know, a very quick thing you're going to be able to see
 4   is whether or not stocks in which you've raised tick sizes
 5   have higher dollar trading volumes, because if you don't,
 6   you're not going to be able to see a whole lot of improvement
 7   in a lot of these measures.
 8             So, those are things to keep in mind, I think, as
 9   we go forward.
10             MR. CLARK:  I think we would naturally expect that,
11   you know, market maker participation would increase
12   proportionally more than institutional posters, and that's
13   the -- that's what you're trying to -- to incent, and if it
14   also does incent institutional posting, I think that would be
15   a great thing, but typically, you know, the -- the
16   institutions are the takers of liquidity.
17             So, to Maureen's point, you would hope that that
18   would result in additional dollar value traded, and that's
19   really one of the key questions:  Does displaying more result
20   in, you know, a different behavior from the institutional
21   customer and draw them in to trade more?
22             Just a couple other, I think, related metrics
23   around, you know, institutions and trading size that could be
24   interesting to look at are how quickly does it take to
25   execute a larger size order?
0177
 1             Typically, in the smaller cap names, you have to be
 2   very patient; it takes a very long time.  So, the hope would
 3   be that this increased liquidity would speed the process at
 4   which someone can get, you know, a sizeable position.
 5             And then the other related metric would be the
 6   transaction cost analysis or what you call the slippage, you
 7   know, of that trade, and what does -- you know, what's the
 8   impact -- impact cost of a large trade, you know, in a market
 9   where there's more depth, and hopefully those are -- those
10   are positive results.
11             MR. SACHS:  I would echo that about the length of
12   time, although arguably it is a difficult metric, I think, to
13   measure and get the information on, but again, from a
14   institutional, you know, manager's perspective who is
15   operating in a market, you know, that is constrained by
16   liquidity, where you're seeking to, you know, take -- buy
17   liquidity for multiple days' worth of average volume, an
18   interesting thing that I have certainly experienced in the
19   post-decimal world, as well as has been reiterated in some
20   research, I think most recently in -- in a piece in 2004 --
21   Bess and Binder, I believe, was the author, if I have my
22   notes correct -- institutional orders that were -- in the
23   post-decimal world -- that were less than one day in length
24   saw increased transaction costs in the decimal world while
25   multi-day orders actually saw a significant decrease.
0178
 1             So, when all of us talk about decreased, you know,
 2   transaction costs from decimals, ultimately my belief had
 3   always been that, while we are certainly seeing a decrease in
 4   cost, you know, we have simply spread out or order times, you
 5   know, to account for the other effects of decimalization,
 6   most namely is the, you know, lacking of resting liquidity,
 7   small order size, or the decline in order size.
 8             So, ultimately, arguably, it's a very, very
 9   difficult metric to get your arms around and to get
10   information on, but I do think it's an important metric that,
11   if it can be measured, to look at trade length or duration in
12   the marketplace relative to these metrics.
13             MR. REED:  One thing to add here is that I think if
14   we start focusing on sort of the wider effects in terms of
15   capital formation, in terms of cost of capital, one of the
16   things that starts to matter a lot is the information
17   environment that these firms are operating in, and there's
18   tremendous disparity in the information environment between
19   very large stocks and very small stocks, and so, it's
20   probably important to take that into account in some way in
21   designing any sort of pilot.
22             You'd want to be able to match carefully on
23   dimensions other than just these direct measures of trading
24   and liquidity but things like news, things like coverage
25   would sort of be at the heart of a lot of the arguments that
0179
 1   were happening this morning.
 2             I think those linkages -- it would be nice to make
 3   sure we knew more about those linkages.
 4             MS. EDWARDS:  Okay.  And that's a good transition
 5   into actually getting into the design elements.  
 6             So, we have had earlier discussions today on some
 7   issues related to design, on which securities should be
 8   included, but I think keeping the goals in mind and what's
 9   feasible to study in mind, I think I'd like to ask the group
10   to opine a bit on how to design it.
11             How should these securities be selected?  I think,
12   how long should the pilot be?  And what should the tick sizes
13   be for these different types of securities?  Should we -- how
14   many different tick sizes do we want to have in a pilot?
15             And I guess adding onto that is what sort of data
16   would be necessary to address the goals of the pilot that
17   we've put forth today?
18             MR. CLARK:  Can I take the easy part of that
19   question?  In terms of the amount of time and number of
20   securities, I think it is important to, you know, have a
21   length of time that's sufficient and a number of securities
22   that's sufficient to motivate changes in behavior.
23             So, from our standpoint, you know, at least a year,
24   and you know, a pretty substantial, you know, group of stocks
25   of 300 to 500 might be needed to motivate the trading
0180
 1   community to make the necessary changes to respond to the
 2   change in tick size.
 3             MS. McCARTHY:  We agree with that.  We think that
 4   it should be at least a year and 500 names.  Again, if there
 5   could be a paired non-pilot group of 500 names, as well, that
 6   would be ideal.
 7             Even though it would be a year or more, even in
 8   that -- even in that instance, we do feel that some behaviors
 9   aren't going to change just by the nature of it being a pilot
10   and being a temporary situation.
11             That is, you may not see the level of research
12   providers improve, if you believe that liquidity and -- and
13   economics -- being incented on an economic basis leads to
14   more research providers in that name, but it takes,
15   typically, three to six months or more to research -- put a
16   research piece out on one name, and it's -- it's unlikely
17   that we're going to change sell-side behavior to incent them
18   to move into this space on a research basis if it's just a
19   temporary, by nature, pilot design.
20             And the same also might be true of liquidity
21   providers and capital commitment in that it might invite
22   those who already have those programs in place to incent them
23   to engage in those practices, but it's unlikely to, in a
24   temporary process, invite new participants.
25             So, that is one thing we would caution.  One of our
0181
 1   thoughts is to add non-statistical surveys to both the buy
 2   side and the sell side halfway through the pilot to determine
 3   whether the buy side sees an impact to what they're doing in
 4   terms of liquidity, and from the sell side perspective,
 5   whether such a program might invite them, once it's
 6   permanent, to engage in further research, capital commitment,
 7   or adding more liquidity.
 8             MR. WOMACK:  On the question of research coverage,
 9   this is only a gut feel, but it's an informed gut feel.  I
10   don't think a year would be even close to enough.  We're
11   talking two or three or four before the market would
12   fundamentally change.
13             You know, there's been a lot of change in research
14   coverage and how it works since the early 2000s when, you
15   know, analyst disclosure came to be.  I'm not at all
16   convinced it's going to change back very much, if ever.
17             You know, we will find a different kind of
18   provision, but I'm not exactly sure that it's going to be
19   brokerage-connected information provisions.  So, I guess what
20   I'm trying to say, in a nutshell, I don't think we'll learn
21   anything in a year with respect to coverage of -- information
22   coverage.
23             MS. O'HARA:  I kind of share Kent's view.  I think
24   it will take quite a while before that could happen.  Within
25   a year, a pilot, though, could give us a lot of information
0182
 1   about changing liquidity production which is somehow related
 2   to all of this.
 3             I would stress, you know, 500 stocks sounds like a
 4   lot, except you're going to have to control for a variety of
 5   things that we know are important.  
 6             For example, market cap, right?  You really need to
 7   think about who is this pilot applying to, and you know, you
 8   might -- you might design like four groups with nano-caps,
 9   micro-caps, very small caps and small caps, so you'd sort of
10   have four groups, and then you also need to recognize that
11   those stocks trade at different price levels, right?  And we
12   know that low-price stocks tend to have a very different type
13   of liquidity process than high-price stocks.
14             So, if you were to divine, you know, four more
15   groups on prices, then you actually now have 16 groups of
16   stocks, if you think about it, and you'd also then have to
17   design, you know, matching control groups.
18             If you pick up what I think is Adam's very good
19   point that information differences across stocks are
20   important, then you may want to think about the fact that
21   perhaps information differences are also, perhaps, not
22   totally, but correlated with industries, right?
23             Certain industries are certainly -- have different
24   information structure.
25             So, you may want to start matching on industries,
0183
 1   as well, in order to really begin to say, when I get to the
 2   end of this pilot, I have controlled for price, I have
 3   controlled for market cap, I have controlled for information,
 4   and now the differences are truly due to the change that I
 5   made in tick size.
 6             If you do that, you need a lot of stocks, and I
 7   completely share the view that, unless it's for a year, the
 8   cost to people to change their systems and keep track of all
 9   this is just too large.  You just won't get people, you know,
10   participating the way you want.
11             So, I mean, it's certainly doable.  Doing match
12   studies like this is something that all of you are very good
13   at.
14             So, it just, though, that it's -- it's not trivial
15   to think about this carefully and plan it in front.
16             MR. LEWIS:  I'd like to add just one thing for you
17   to consider, as well, from the perspective of someone who has
18   actually spent a fair amount of time thinking about how to
19   design pilot programs, and that is, when you have a -- I
20   asked everybody for what they thought the market failure was,
21   right, and we heard a number of different alternative
22   interpretations that characterize what we should be trying to
23   look at in a pilot study, and I understand that you can have
24   a very focused pilot study that takes on one of those
25   concerns; you could have a more involved, complicated, pilot
0184
 1   study that's simultaneously designed to try to address a
 2   number of these different issues.
 3             But bear in mind that every time you bring a new
 4   issue on the table, you're effectively constructing a brand
 5   new sample.
 6             Now, that also has to be considered against the
 7   backdrop of the fact that, at the Commission, we have a
 8   number of other pilot studies that are already underway, and
 9   so, when you start to think about how many available
10   securities there are to use for a pilot study, you have to
11   worry about the possibility of cross contamination, that
12   what's being controlled for by using it in one pilot study
13   has to be considered how it affects an alternative pilot
14   study.
15             So, it's important to keep the focus precise, so
16   that you understand exactly the nature of the question you're
17   trying to answer.
18             MR. JIGANTI:  Maureen hit it right on the head. 
19   With the 16 different combinations -- and I would actually go
20   to 32 and add a couple of price points in there.  You know,
21   we've avoided the number.
22             I think a two-cent pilot in some instances makes
23   sense and a nickel pilot, both, you know -- so, if we had
24   your 16, we spread them over those -- turn them into 32 --
25   could end up with a pilot program with -- with real data,
0185
 1   with price point differences, to measure against each other
 2   in the same market conditions, which brings me to -- Kent is
 3   absolutely correct.  A year isn't long enough.
 4             The liquidity providers will jump in one day one if
 5   they see an opportunity that makes it work, but the -- but if
 6   you want to get more research people in -- I don't know how
 7   many people here have hired anyone recently, but it doesn't
 8   take -- it doesn't happen in a day, and so, I imagine that
 9   it's going to -- it's going to keep -- it's going to keep the
10   doors closed for a while.
11             So, if you're looking to see if the research
12   departments get to be greater, it's got to be a minimum of
13   two years, in my mind, before you start seeing any real
14   change out there.
15             MR. REED:  Just to add one thing, the last panel
16   was concerned about the idea of opting in.  If we had an
17   opt-in feature to the pilot, that reduces the sample size
18   significantly there and makes all these concerns even more
19   problematic.
20             One other thing I just wanted to bring up -- I
21   don't think I have a good solution to this, but if one of the
22   goals of the pilot is to think about capital formation, then
23   effectively you're trying to run a pilot for companies that
24   have not yet entered the market, and so, you'd need some way
25   of exogenously determining which firms would be allowed to
0186
 1   have high tick sizes and low tick sizes, and it seems like it
 2   would raise a lot of procedural issues for the SEC.
 3             MR. HATHEWAY:  Just to offer my two cents on this,
 4   I don't think we'd necessarily need 32 partitions of
 5   potential sample, taking everything outside the Russell 3000,
 6   taking every nth stock is a simple enough way, and on a
 7   mixing perspective, you should get a lot of the various
 8   attributes you're looking to cover covered.
 9             I do think Paul has a very good point, that there
10   should not be one tick for the pilot.  Having multiple ticks
11   within the pilot stock allows studies to be done when
12   individual stocks cross price points.  It allows us to get
13   some familiarity with what calibrations may be, and it also
14   avoids placing one-size-fits-all regime with a
15   two-possible-sizes-fits-all regime, which is not necessarily
16   all that better.
17             I think ideally, at a minimum, we need to do, as I
18   said, the stocks sort of outside the Russell 3000.  There's
19   no index issue.  These are essentially stocks that are not
20   the most liquid in the world.
21             Ideally, we would take a look at larger stocks, as
22   well.  There are investor costs associated with a tick too
23   wide.  I've been in here in the past before Citi split to
24   point out that essentially professional money was trading on
25   a half-cent spread and non-professional money was trading a
0187
 1   few tenths of a cent or hundredths of a cent inside the full
 2   penny spread, and that's not necessarily clear that the
 3   individual investor, in cases like that, is getting the best
 4   possible deal.
 5             MR. LEWIS:  It's a very naïve question, but I'm
 6   wondering whether issuers and/or investors might put up a
 7   stink about us allocating them to two cents versus five
 8   versus 10.  I mean, can we imagine that they would not like
 9   the --
10             MR. JIGANTI:  It can't be as bad as the short sale
11   pilot.  I don't think many -- you guys have to tell me, but I
12   have a feeling there were firms who were not happy about
13   losing their tick test for a study.
14             MS. EDWARDS:  One quick question before we move on
15   to costs and benefits, and we'll probably address your point
16   again, Kent, in talking about costs and benefits, but there's
17   several ways that a pilot can occur.  
18             Either the SEC can do a pilot such as the
19   Regulation SHO pilot or the exchanges themselves can do a
20   pilot through an NMS plan such as limit-up/limit-down pilot.
21             Which way would be the best way to go forward if a
22   tick test pilot was appropriate?
23             MR. HATHEWAY:  Plans are becoming increasingly
24   popular.  I suppose you may have -- it's a way to take
25   commission, share scarce resources, and leverage them in
0188
 1   terms of getting things designed and implemented, I'm just
 2   guessing.
 3             I would guess, with a plan, there's probably a
 4   little bit longer time to get it done.  You've got a whole
 5   bunch of SROs in the room who have to agree on something. 
 6   Presumably people other than the SROs would be represented on
 7   advisory boards like we do with the data plans.
 8             I've thought the same question, and I think, in
 9   this case, given what we're already asking the SROs to do
10   with other plans, and we're getting some experience in
11   working with each other on other plans, and less
12   contentiously than recent history with the data plans, this
13   may well be a good place to try a plan.
14             If we can get limit-up/limit-down done with a plan
15   -- we haven't got there yet, but if we should, then this
16   would be something we could do with a plan, though, again, we
17   haven't discussed that internally, so I'm speaking mostly for
18   me.
19             MS. SEIDEL:  Before we leave the design, I just had
20   one, further question for the panel.  There's been a lot of
21   discussion about how we might decide what securities would go
22   into a pilot or have a pilot, or the goals that we might want
23   to measure during a pilot.  Were we to consider a pilot,
24   should we be giving any consideration now, at this point, as
25   to sort of is there anything thought we should give to
0189
 1   whether or under what circumstances securities would come out
 2   of a pilot?  Or, is that something we should think about when
 3   we're thinking about the design of the pilot from the start;
 4   or, once we're ready to have a pilot, once we actually got to
 5   see some data, see what's going on?  Just sort of interested
 6   in views on that.
 7             MS.  MCCARTHY:  Are you talking about an opt out or
 8   are you talking about, now, the determining when this --
 9             MS. SEIDEL:  Sort of whether or not we should be
10   giving thought --
11             MS. MCCARTHY:  They pass the metric.
12             MS. SEIDEL:  -- right.  Is there some parameters,
13   what might be the parameters, at what point consideration
14   should be given.  Should there be, you know, for some
15   standards in place where securities will come out of the
16   pilot.
17             MS. MCCARTHY:  Right.  Our feeling is, I guess,
18   I'll answer both questions.  Our feeling is for after a
19   statistically significant period of time that companies
20   should be allowed to opt out for whatever reason, if they
21   determine it's harmful to them in any way.  But, separately,
22   I think you're also asking if it moves out of a market cap
23   metric into another one, what's the situation there.  I think
24   it depends on the time design of the trial.  If it's a year,
25   no.  If it's a much longer study than a year, then I think
0190
 1   one has to determine what happens once they moved
 2   significantly out of the metrics.  They got them into the
 3   trial in the first place.  
 4             MR. SACHS:  And, again, I would echo that. 
 5   Theoretically, I can't quite get my head around the opt in,
 6   opt out, and how that would actually affect a study like
 7   this.  I think there's too many variables to control for
 8   there, but, ultimately, yes.  I think if you're going to move
 9   stocks out of the pilot program, I think very
10   straightforward, quantitative, simplistic approach of market
11   capitalization and/or something else, as few variables as
12   possible, I think, to move something out of the program.  
13             I think, depending on what the length of the time
14   of the program is -- and I do think longer in this particular
15   program is better than shorter -- I think you're going to run
16   into issues of the data being subject to too much analysis if
17   you have a large number of stocks moving out of the program.
18             And, then, what replaces those?  What sort of
19   market regime are you in when those stocks come in and go
20   out, because everything we've talked about here from a design
21   perspective is extremely important, not the least of which is
22   the industry or sector the security is in.  Because, again,
23   depending on the length of time -- but over a multi-year
24   period, obviously -- macro economic regimes are going to
25   change and certain stocks are going to start behaving
0191
 1   differently as flows move towards those stocks and then flows
 2   move out of those stocks, or, you know, the sectors.  Money
 3   rotates through the sectors and out of the sectors.
 4             MR. JIGANTI:  I think that if the stock -- one way
 5   out would be to have the stock at the S&P 500.  Some major
 6   event would be an obvious opt out that would have enough
 7   natural liquidity that would come from index funds that would
 8   be trading.  That becomes more of an every-day stock.  That
 9   would be one way out.  The other is a defined end to the
10   pilot program -- you know -- if it's a three-year program
11   that is actually a three-year program and it ends with a set
12   date.
13             MR. REED:  I would just add that from the
14   perspective of pure research the opt in and opt out are
15   similar issues and they have sort of androgyneity.  You'd be
16   concerned about the particular type of firm choosing the opt
17   out, because of its circumstances.  You wouldn't necessarily
18   be able to tell the effect of the pilot study if there was
19   this ability to opt out.
20             The other thing I'll just mention is that depending
21   on the fixed criteria that you use to allow a firm to
22   graduate or to move out of the pilot, you'd want to be
23   careful about that, and just allow for the possibility that
24   some metrics might be gained in order to effectively opt out. 
25   So you'd just want to design those in ways that are difficult
0192
 1   to challenge them as exogenous as possible.  
 2             MS. HANLEY:  So I wanted to ask a question about
 3   potentially a non-pilot.  So does any one thing that perhaps
 4   I've heard that issuers -- issuers in the last count --
 5   issuers would be happy to choose and take sides.  Would it be
 6   perhaps better to go to a regime in which instead of having a
 7   pilot we allowed companies to choose their own tick size and
 8   do it that way?
 9             MS. MCCARTHY:  I guess I think two things there. 
10   One is I think I like the competitive, natural selection
11   there; however, I do think we're asking companies to consider
12   something about which they know less than we do on, which is
13   market structure, and the nuance is a market structure.  I
14   think -- we think we would be better if we told companies
15   whether they are in or out, but as I said, I do like the
16   natural selection of that.  So that's a tough one.
17             MR. HATHEWAY:  Yeah.  While issuer choice is very
18   appealing, I do echo Maureen's concern that it's not a lot of
19   information the firms would have for the basis of making a
20   decision.  It may well be to an outcome where firm's just do
21   what everybody else does, and we don't really gain much
22   information.  Issuer choice may be a possible end state,
23   particularly if the results are sort of it depends, if we
24   actually do a pilot.  But I'm not sure if we would be able to
25   gain information from issuer choice, if that were the pilot
0193
 1   itself. 
 2             MR. CLARK:  The issuer choice exists today.  I
 3   mean, right, so all issuers can choose now to reduce their
 4   stock price to effectively widen their spread and, I guess,
 5   it's an option that exists today.  We could try to educate
 6   them better on that process; but, ultimately, what you're
 7   going to end up doing is just every one of these companies
 8   will be trading at below $10.  And is that the outcome that
 9   we're looking for?
10             MS. HANLEY:  Can we move now to some costs
11   associated with the pilot?  So, operationally, those of you
12   who would be involved in this, operationally, what would be
13   the cost this pilot?  Would it be difficult to implement a
14   tick size change?
15             MR. CLARK:  So I've talked to our technology guys,
16   because this is essentially something we do in Europe
17   managing multiple ticks, and it's not that different
18   technology between U.S. and Europe.  From our perspective as
19   an exchange operator, it's not a particularly costly
20   implementation.  I think the bigger costs will be probably
21   borne by the participant firms, particularly those that have
22   a large client face that's not used to seeing anything other
23   than pennies.
24             MR. JIGANTI:  It's a good segue into we can make
25   the change.  When options are traded in multiple different
0194
 1   price levels, it was confusing going in.  It's really more of
 2   an educational process cost than the actual program costs;
 3   you know.  Every time you make those minor changes, there's
 4   obviously some cost to the industry of potential hazards of
 5   changing code.  But since we do it every day, I think,
 6   ultimately, that's not a very large cost.  It's really
 7   education.
 8             MR. SACHS:  From the investment management
 9   perspective, I see little to no significant costs.  You know;
10   obviously, from an operational perspective, most of that
11   would be borne by data providers, broker-dealers, exchanges. 
12   Certainly, there would be costs from a time and resource
13   perspective in transaction cost analysis and reporting, and
14   the way in which we look at transactions in stocks in the
15   pilot program; but I would suspect that those costs would be
16   minimal. 
17             MS. HANLEY:  What about other types of costs, so
18   unintended consequences.  There can be, as Maureen pointed,
19   sometimes winners and losers in this.  And so how do we
20   balance that?  I know we had some experience in the reg show
21   pilot but how do we balance that, particularly in stocks that
22   are what we would call, some have said here, fragile.  
23             The eco system generally, we're only changing one
24   aspect of the eco system.  We're not changing the many things
25   that people have talked about here, and that these stocks are
0195
 1   already facing concerns about trading that the effect on them
 2   may be greater, let us say, than if they did this to a very
 3   large company.  So if you could talk a little bit about some
 4   of the potential, unintended consequences that could occur,
 5   or areas which we're particularly mindful, so that we don't
 6   further harm a market that others have suggested as
 7   struggling.  
 8             MR. HATHEWAY:  Well, quoting algorithms -- and
 9   everybody now quotes algorithms now -- are expensive to cope. 
10   There's not a lot of trading volume, and some of the stocks
11   we're talking about.  And one potential outcome is you lose
12   liquidity providers, because they just choose not to recode
13   their systems.  And that's a cost.  It damages the pilot and
14   potentially, negatively impacts investors if you get enough
15   liquidity providers opting out.
16             There's certainly going to be some investors who
17   will pay more.  I mean averages are averages.  There are
18   going to be some investors who pay less.  The goal is that
19   investors pay less and they have better access to small
20   growth companies.  That's harder to quantify, and the welfare
21   arguments around who's paying more and who's paying less is
22   particularly difficult to quantify.
23             I mean one suggestion I made on an earlier panel
24   was that if you narrow ticks in some stocks, investors as a
25   class will be receiving the benefits from presumably higher
0196
 1   volume stocks with a narrower tick and incurring some of the
 2   costs associated with less active stocks with water ticks. 
 3   Clearly, there were members of some of the earlier panels who
 4   were not receptive to narrower ticks and active securities.
 5             MR. NALLENGARA:  We've identified the market
 6   failure, or at least we've talked about it; but this panel
 7   was designed to talk about a pilot.  But are there other --
 8   as we consider it and we look at costs and benefits -- we'd
 9   also consider what other alternatives there would be?  Are
10   there other market incentives that could be, should be
11   considered other than a pilot that any of you have thought
12   about?  I'm just wondering if there's something other than
13   ticks that should be looked at in considering this market
14   failure.  
15             MR. CLARK:  I think this is going to be a very long
16   panel if we go there, but I think that, certainly, we're
17   cognizant that there are a lot of issues in the marketplace,
18   and this is not a miracle cure to resolve the liquidity
19   challenges of the small cap market.  I think in conjunction
20   with this pilot, we need to continue to move forward with a
21   holistic review of overall market structure, various
22   dimensions of it, which also could improve the environment in
23   a less active securities as well as more active securities. 
24   Obviously, there's a lot of topics out there related to
25   maker-taker models related to fragmentation, related to other
0197
 1   types of, maybe, market maker incentive programs.  You know. 
 2   We're working with the Commission now on a program in the ETF
 3   space where the issuer is providing payment to the market
 4   maker to meet certain performance obligations.  And I think
 5   that's an interesting pilot.  Well, it's not a pilot yet, but
 6   it has interesting potential; you know, if the results prove
 7   interesting there.  You know.  That's something that could be
 8   extended into the smaller cap space, for example.
 9             MR. SACHS:  Yeah.  I would echo that.  I mean,
10   obviously, stating the obvious here, I mean, really again it
11   depends on what problem we're trying to solve.  You know.  If
12   you're looking at the lack of IPO activity and/or the
13   hampering of the research efforts from the broker-dealer
14   community, I think from a market structure perspective there
15   are far less things that you could do or leverage you could
16   pull to have immediate impact.
17             If you're looking solely at liquidity and the
18   quality of market, if you will, and measuring that by some of
19   the metrics that we've spoken about here earlier, depth of
20   book, average size of trade, you know, overall liquidity
21   perspectives, then yes, I do think that there's probably a
22   fairly long list of things that have been discussed at length
23   from the maker-taker model to market maker obligations, all
24   sorts of things.
25             But, again, to echo what was just said, I do argue
0198
 1   that if we have the ability to take a step back and look at
 2   the entire market structure eco system as opposed to looking
 3   at individual items that are on the list -- such as spreads
 4   and small cap stocks and the maker-taker -- and really look
 5   at it holistically, obviously, understanding the realities of
 6   the world we live in, I do think we ultimately would end up
 7   in a better place from a market structure perspective as
 8   opposed to handling the situations in a piecemeal basis to
 9   react to certain perceived disruptions or failures in the
10   marketplace.  
11             MR. WOMACK:  Professor Jay Ritter from Florida
12   wrote a very wise article that I think was showcased here a
13   few months ago about re-energizing the IPO markets.  And I
14   think I would encourage you to go back and think about some
15   of the issues that he brought up; number one, reducing the
16   overall costs of going public.  You know.  This is a market
17   where despite contraction in secondary costs, the primary
18   cost of going public is still very large.  You're talking 6
19   to 7% plus, typically 10 to 15% for the best deals.  
20             That's a very large cost to the issuer, and I think
21   that belongs on the table too, if we are thinking about what
22   energizes investors and issuers to want to be back in this
23   IPO market again.  So I wouldn't forget to go back and look
24   at some of the things that Jay twisted our arm about.
25             MR. REED:  Also, and in some of Jay Ritter's slides
0199
 1   related to the same paper, he mentions this idea of research
 2   being paid for by the issuing firms.  I think we were talking
 3   here about complicated linkage between tick sizes and
 4   eventually the information environment in which a firm
 5   operates.  I think if the firm really does value that
 6   information environment, perhaps we should at least consider
 7   mechanisms that allow firms to pay for that information
 8   directly, instead of hoping that market maker profits will
 9   generate research and that that will generate liquidity.
10             MR. BERMAN:  Can I ask a follow-up question based
11   on the theme that Lona was picking up on and I think that,
12   Stephen, you and a few others were reacting to?  The market
13   is very complicated.  People have talked about maker-taker. 
14   Both here and also in both of the first two panels, there was
15   also discussions of off-market trading.  So are there things
16   that we need to do for market structure purposes to control
17   for in any sort of a pilot?  
18             So where I was thinking is if we're looking for
19   improved liquidity, improved depth of book, et cetera, let's
20   assume that we find trading volumes -- average trading volume
21   -- goes up.  But it's actually 70% is now off exchange, and
22   the depth of book actually has gone down.  What would you
23   conclude from a pilot?  Would you say, well, that's actually
24   good?  Or, actually, is that bad, because we were looking for
25   depth of book?  So are there linkages between off-market
0200
 1   trading or over-the-counter trading that we need to control
 2   for for the purpose of understanding the pilot?
 3             MR. JIGANTI:  If that's not real market -- if it
 4   goes on the other parts of the marketplace -- if you take one
 5   of those off-market trading pieces away, you will affect the
 6   pilot in ways that you may not even realize.  You know.  So
 7   many of these things we want to see how they play out in real
 8   life if we're going to think about and impose non-realistic
 9   limitations to it.  You're not going to have the same result
10   that you would.  You know; if it turns into a market
11   structure change, it's a different pilot.
12             MR. SACHS:  But, again, from the practitioner's
13   perspective, I would argue that the reason that volume moves
14   elsewhere off exchange is in reaction to the market structure
15   in which we're forced to deal with today -- lack of
16   liquidity, flickering quotes, you know, noise if you will,
17   generated by the market structure that we're in.  So I would
18   argue, while I'm guessing it wouldn't be at all popular,
19   controlling for all those factors and only allowing for, say,
20   true mid-point, dark crossing at the middle of whatever the
21   pilot spread is.  
22             You know, it eliminated control for many of those
23   factors, so you would still allow for, again, large orders to
24   cross in a dark pool or what have you in the middle of the
25   spread, but you wouldn't eliminate all of the other sort of
0201
 1   ancillary noise that you get from small price improvement --
 2   you know, tenths of a penny, that sort of thing.  And, again,
 3   obviously, each constituent represented here, whether it be
 4   the exchange or the broker-dealer community, or the
 5   investment management community, it's obviously going to have
 6   a greatly different opinion of that.  
 7             MR. HATHEWAY:  Yeah.  I think, well, two responses: 
 8   First place, that's one of the things a control does, that
 9   controls stocks, is lets you get a sense of what's happening
10   elsewhere in the market for stocks not in the pilot; the
11   other thing, the goal is depth.  The goal is to improve
12   liquidity.  The goal is to lower cost of capital and a more
13   attractive environment for small companies.  If a larger,
14   wider display tick were to result in all these other things
15   happening, then it's probably a good income.  The locus of
16   trading, per se, and competitive dynamic between NASDAQ, NYC,
17   Broker Systems, BATS, Direct Hedge, you know, their
18   exchanges, is probably not a critical point.  So if a wider,
19   not particularly used displayed quote allows people to have a
20   better sense of where the price really is to get size done,
21   to get retailers done more effectively, then all in all
22   that's a good outcome.  
23             And speaking now as an exchange rep, we would be
24   back here saying, hey, what these guys are doing in the dark
25   is the right way to trade with inside a displayed quote, and
0202
 1   we want to replicate those mechanisms.  But taking a step
 2   back from that, the scenario you described could be good or
 3   ill, depending on the other attributes you've already talked
 4   about.
 5             MS. MCCARTHY:  Yeah.  I agree with that.  I think
 6   that if you were to do what Frank said and you allowed
 7   trading, just allow trading at the midpoint, you get around a
 8   few things, which is the dark pools can still participate. 
 9   The opening cross would be critical for that, however, I do
10   think you're starting to add some elements and maybe getting
11   away from what is the end point, and maybe that's a little
12   tougher to design.  
13             If you were to say that nothing could be traded in
14   the middle, I don't think that -- now I think we're into a
15   completely different study.  And I'm not arguing that it
16   shouldn't be done.  I'm just saying I don't know if it's in
17   -- well, without becoming a completely different trial design
18   here, and I guess that's it.  
19             MR. CLARK:  And I just add to that.  I think the
20   study is to explore the impact of changes in tick sizes; and,
21   certainly, there should be a control group that's just
22   focusing on that change.  If you didn't want to add other
23   elements to it, you could have another separate control group
24   that's looking at tick size change, plus elimination of some
25   pennies, or whatever that other variable is you want to look
0203
 1   at.  And that way you could look at the data to see what the
 2   impact of that element is; but I think you don't want to lose
 3   sight of -- just focus on -- the tick size in and of itself.  
 4             MR. BURNS:  I guess the question I had, which is
 5   quite naive, is if we were to put rules in place for a pilot,
 6   is it possible that investors and/or market makers can just
 7   reverse engineer around it and make it a bit null and void in
 8   terms of what our true goal is?  I guess Frankie or somebody
 9   else would be far better at thinking about such things than
10   I.
11             MR. LEWIS:  So, Frank, could I jump in here real
12   quick and sort of add something on top of that?  And it goes
13   back to the point I think Greg was making.  And, that is, if
14   you were going to design a pilot, and we have to consider the
15   unintended consequences of the pilot, we ought to consider
16   where order flow is likely to go.  Right?  So discussion now
17   is assume a static framework where there is no movement.  
18             Will transactions continue to be conducted in
19   markets at the same rate, or will people be incentivized to
20   actually take their trades into dark pools?  And I think that
21   goes to the arbitrage question that Kent has raised.
22             MR. HATHEWAY:  In illiquid stocks, roughly, half
23   the volume is probably going to be in dark pools already, our
24   internal safety platforms of one type or another.  So could
25   you move more?  Yeah.  Of course, it's still not 100.  But to
0204
 1   Kent's point about "would people try to reverse engineer it,"
 2   yeah.  They would.  This is one of the fascinating things
 3   about this industry.  You show people where lines are and the
 4   immediately start proving.  Exactly where the line in fact
 5   is, through all the steps the SEC took regs show, there were
 6   things that people came up with to get her on locate, to get
 7   her on some of the other things that people did.  
 8             And I think Greg's show is many respects
 9   illustrative of when the Commission has a commitment to a
10   goal and takes step by step processes to get it done, they
11   get it done.  It may not be, sort of, day one; but, you guys
12   would really need it to be on making sure it's selling and
13   similar issues underlying reg show.  Would people reverse
14   engineer?  Yes.  Would there be a change in the dynamic of
15   how off exchange trading operates?  Yes.  How is that going
16   to come down?  These are not profitable stocks to trade.  
17             That's the bottom line on these securities.  So
18   when we were talking about the high frequency trading firms,
19   they're not active in these securities.  It's essentially
20   various people's algorithms trying to get a trade done or
21   trying to get somebody to get a trade done at a price they
22   may not like to expost.  Executors were hearing about the
23   forced asset share trader earlier this morning.  Is it going
24   to change?  Yeah.  But I think the whole point of a pilot is
25   to find out what net-net we accomplish in terms of creating a
0205
 1   more attractive ecosystem for companies whose behavior over
 2   the last 12 years is clearly indicated; but, public equity is
 3   just not a very attractive place for them to be.  
 4             And that, as just pointed out in the earlier
 5   panels, has implications for entrepreneurship.  It has
 6   implications for the type of companies to get started.  It
 7   has implications for the ability of investors to participate
 8   in potential, high growth companies.  Risky, yes, but with
 9   these companies not being a part of the companies now, you're
10   closing investors off to the opportunities to participate in
11   them.  So I think it's a learn from a pilot, Maureen's point
12   about seeing more than the displayed information for the
13   markets is well taken.  
14             What are people doing that doesn't necessarily get
15   displayed in our quotes is a relevant part of this today that
16   really wasn't that big a part of the story 12 years ago.  The
17   world was different when we did decimalization.  
18             MR. BURNS:  Well, irrespective of where we go with
19   the pilot, I hope the stringers were all listening when Frank
20   said we were a "get it done" kind of agency.  And, once
21   again, like the technology roundtable this past fall -- for
22   my part, parochial interest -- I think the quality of the
23   panelists and the discussion today was superb.  We thank you
24   so much for the time you've taken with us; and, doubtless,
25   you and others will be coming and visiting and making
0206
 1   submissions, which we welcome very much in coming weeks.  
 2             Thank you very much and thank you to my colleagues,
 3   the staff here, the Commissioners for permitting us to hold
 4   this roundtable, and for all of you.  And on behalf of Greg
 5   and Lana, and all of us here, thank you very much.
 6             (Whereupon, at 4:03 p.m., the roundtable was
 7   concluded.)
 8                           *  *  *  *  *
 9   
10   
11   
12   
13   
14   
15   
16   
17   
18   
19   
20   
21   
22   
23   
24   
25   
0207
 1                  PROOFREADER'S CERTIFICATE
 2   
 3   In the Matter of:   ROUNDTABLE ON DECIMALIZATION
 4   File Number:        OS-4-657
 5   Date:               Tuesday, February 5, 2012
 6   Location:           Washington, D.C.
 7   
 8   
 9        This is to certify that I, Maria E. Paulsen, 
10   (the undersigned), do hereby swear and affirm 
11   that the attached proceedings before the U.S. 
12   Securities and Exchange Commission were held 
13   according to the record and that this is the 
14   original, complete, true and accurate transcript 
15   that has been compared to the reporting or recording 
16   accomplished at the hearing.
17   
18   
19   
20   ____________________     ____________________
21   (Proofreader's Name)           (Date)
22   
23   
24   
25   
0208
 1                   REPORTER'S CERTIFICATE
 2   
 3   
 4   I, Jon Hundley, reporter, hereby certify that the 
 5   foregoing transcript of 206 pages is a complete, true and
 6   accurate transcript of the testimony indicated, held on
 7   February 5, 2013, at Washington, D.C. in the matter of:  
 8   ROUNDTABLE ON DECIMALIZATION.
 9   
10   
11   I further certify that this proceeding was recorded by me,
12   and that the foregoing transcript has been prepared under my
13   direction.
14   
15   
16   
17                                Date: __________________________
18                   Official Reporter: __________________________
19                   Diversified Reporting Services, Inc.         
20   
21   
22   
23   
24   
25   
0209
 1   
 2   
 3   
 4   Diversified Reporting Services, Inc.
 5   1101 Sixteenth Street, N.W.
 6   2nd Floor
 7   Washington, DC  20036
 8   
 9   
10   In the Matter of:   ROUNDTABLE ON DECIMALIZATION
11   File Number:        OS-4-657
12   Date:               Tuesday, February 5, 2012
13   Location:           Washington, D.C.
14   
15   This is a letter to inform you that we do not
16   release our tapes and notes.  I do maintain
17   them for a period of one (1) year.
18   
19   Sincerely,
20                       _________________________
21                       
22   
23   
24   
25