U.S. Securities & Exchange Commission
SEC Seal
Home | Previous Page
U.S. Securities and Exchange Commission

Panel 1 Transcript: Market Structure Roundtable

June 2, 2010

Chairman Mary Schapiro:
Good morning. Welcome to the Securities and Exchange Commission Market Structure Roundtable. We are grateful that so many respected and experienced professionals have agreed to participate as panelists in today’s meeting. And we are looking forward to an instructive day listening to these experts’ comments, insights, and recommendations on critical market structure issues, particularly in the areas of high frequency trading and undisplayed liquidity and the appropriate metrics for evaluating market structure performance. We announced this roundtable two months ago as part of our effort to build a comprehensive response to the extraordinary changes in the structure of our equity markets in the recent years. This effort began back in January with a market structure concept release which has generated valuable commentary that will help shape today’s discussion.

Our roundtable has also been formed by a more recent event, however: the severe, albeit brief, market disruption of May 6. For 20 minutes that afternoon, U.S. financial markets failed to execute their essential price discovery function, experiencing a decline and recovery that was unprecedented in its speed and scope. That period of fluctuating prices both directly harmed investors who traded based on flawed price discovery signals and undermined investors’ faith in the integrity and fairness of our markets. We are making progress on our ongoing review of May 6, and we have begun putting in place safeguards that we expect will help reduce the likelihood of this kind of unusual trading activity from recurring, publishing for comment exchange rules which would establish a five-minute trading pause for certain individual stocks if their prices decline by 10 percent over a five-minute period. But, to be clear, this is very much a first step in what I hope to be the Commission’s response to flaws exposed by the May 6 aberrant trading. But even prior to May 6, the Commission had already launched initiatives to strengthen the U.S. equity markets and to protect investors. We proposed rules that would prohibit flash orders, increase the transparency of dark pools of liquidity, prohibit broker-dealers from providing unfiltered access to exchanges and to create a large trader reporting system. And then just last week we published for comment a proposal for development of a consolidated audit trail. And, as I noted a moment ago, the Commission issued a concept release on market structure in January, soliciting public comments on the impact of different trading venues, strategies and tools, including high frequency trading, on our markets and investors. These issues are especially timely in light of the May 6 events and will be the center of today’s market structure discussion.

The concept release raised questions across three broad categories. First, it asked about the performance of the market structure in recent years, particularly from the standpoint of long term investors. Second, it sought comments on the strategies and tools used by high frequency traders, such as co-location services. And third, it asked about dark liquidity in all its forms, including dark pools, ATS’, OTC market makers and undisplayed order types on exchanges and ECNs. Since this release we have received over 170 comment letters, and we hope to explore many of the important questions and ideas raised in those letters today. During the course of the day, we will hear from three different panels. Each will begin with a panelist taking a few moments to share their thoughts on the topic being discussed. And then the floor will be opened to questions from the moderators and commissioners.

The first panel will provide a partial overview of the current market structure performance and focus on price volatility as the panelists discuss how well the current structure establishes prices and allows investors to efficiently buy and sell stocks both in normal and, importantly, in stressed trading conditions; what sort of cost venue pausing mechanisms do our markets need and how would these mechanisms better serve investors; and what the most useful metrics for assessing market performance should be. In addition, the panel will discuss regulator measures that the Commission might take to improve current market structure and appropriately minimize short term volatility and resulting harm to long term investors.

The second panel will explore the growth of high frequency trading and its effect on the market, including the effects on liquidity and spreads, both in normal and stressed conditions, market impact costs for other market participants, and strategies and tools that are beneficial or harmful to the markets or certain participants. The panel will also ask whether high frequency strategies provide liquidity to the market in a manner comparable to the traditional obligations of market makers, whether high frequency traders enjoy an unfair advantage in the markets, and whether high frequency traders should be subject to any trading obligations that are comparable to those of traditional market makers.

The third panel will discuss issues surrounding undisplayed liquidity, including what liquidity is and what role it plays in today’s markets, access to and limitations on access to different types of dark pools and the market participants that they serve. In addition, we are interested in hearing what portion of institutional and retail order flow is executed in undisplayed markets and whether institutional and retail investors receive better quality executions in dark markets. The panel will also explore whether the volume of trading in the undisplayed venues has become sufficiently large that it is detracting from the quality of price discovery in the public markets, and if so, what regulator responses might be appropriate.

There is a hugely important national conversation taking place about our markets, who they serve and how they accomplish their critical capital formation and price discovery functions. I hope that today’s presentations will help inform us in taking the right steps to insure the integrity of markets that are critical to our economic success. Our panels today are leaders in their respective fields and represent a range of constituencies, including retail and institutional investors, issuers, exchanges, alternative trading systems, financial services firms, high frequency traders, and the academic community. We are privileged to have them here and grateful for the effort that they have made to take part. We look forward to a spirited and substantive discussion.

And now I would like to invite my colleagues to offer their own words of welcome. And then, after they speak, we will turn the proceedings over to Robert Cook, the Director of the Division of Trading and Markets, and his colleagues who will introduce and moderate the first panel.

Commissioner Casey.

Commissioner Kathleen Casey:
Thank you, Chairman Schapiro. I have no other formal remarks other than to join you in welcoming our panelists today, thank them for taking the time, for contributing to what I hope will be a very informative discussion.

Commissioner Elisse Walter:
And I thank you, too, Chairman Schapiro for hosting this event. I am very pleased to be here. We have an exciting and packed day ahead of us. So I will just make a few brief remarks. I would like to start by thanking the very hard working staff in the division of trading and markets for all their hard work in recent months. In preparing our concept release on equity market structure, reviewing the numerous comments to that release, and organizing today’s roundtable. And that’s what they do between 3:00 and 4:00 in the afternoon. You have set forth a robust agenda that touches on many important and timely topics and assembled an impressive group of panelists who represent a wide range of viewpoints. We are sure to have a fruitful discussion. I’d also like to thank all of the panelists, those who are on the stage right now and those who will come later. We truly value your expertise and your unique perspectives and appreciate your willingness to dedicate your time to this event.

As I have previously stated, one of the great strengths of our agency is its willingness to continue to review its regulatory approach and to seek input from the public and, importantly, the industry. This roundtable is evidence of that commitment. A public discussion of these important issues is educational for the general public, provides industry and other represented groups a forum to publicly express their views on this topic, and helps us to zero in on the relevant regulatory initiatives that we need to take to improve the current equity market structure.

I look forward to hearing the diverse viewpoints on all the topics on the agenda. And I will be particularly interested in hearing the panelist’s remarks in response to a few questions. First, how does the duty of best execution affect the way market orders are handled? Secondly, should time out mechanisms for individual stocks be uniform across all trading venues? And alternatively, what are the effects on the rest of the market when only one trading venue implements a particular time out mechanism? Third, is there any place at all for stub quotes in today’s markets, and if not, are there other additional steps that we need to take to deal with the problem stub quotes seem to evidence?

Thank you so much for being here and I am looking forward to the rest of the day.

Commissioner Aguilar:
Thank you. I firmly support the Commission’s effort to evaluate the structure of the capital markets. As with any such effort, the parameters of the inquiry and the choices of who participates in the process will heavily influence the outcome. It is the Commission’s responsibility to objectively collect information from all sources, so that we, the government regulator, can make assessments about the optimum market structure that results in fair and orderly markets. Because of the importance of this goal, I am concerned that the views will hear today at the Commission-sponsored roundtable will overwhelmingly be from representatives of the financial services industry. I want to be clear: I am grateful to all of the panelists for being here throughout the day to share their views, and I understand that their viewpoints are not identical. All of you have important information to share with us about how your businesses work, and I thank you for taking the time to be with us today.

However, I am disappointed that our roundtable is not constituted to showcase the full breadth of relevant voices, particularly from a number of issuers and investors, including retail investors. And I am concerned that as a result today’s discussion will not bring to light how conflicts of interest and particular business models may impact the various views that we will hear today. As regulators, we are seeking information to understand the current structure of the markets and to make sure they are aligned with their proper purpose, serving long term investors and serving companies that need to raise capital so that they can develop products, build plants, and put people to work. I hope and expect that our staff will seek out diverse information and views as this inquiry moves forward. And I hope that such efforts occur in a public forum such as the one today. After all, the American public needs to be assured that the foxes are not the primary consultants in how to build the hen house. We must hear from all points of views to implement a market structure that is fair and orderly and that benefits the American public. As today’s discussion unfolds, I will look forward to a discussion that focuses on the needs of issuers and ordinary investors. And I thank you again for being here.

Commissioner Troy Paredes:
And I just want to extend my thanks to the panelists who are on the stage and those who will be here shortly, for whoever are taking time out of their busy schedules to be here and to help inform our process. And, of course, as always, to thank the staff in the Division of Trading and Markets for all of their efforts, not only pulling this roundtable together, which takes an incredible amount of work, but for all that they have been doing leading up to this point and all that they will be doing after today’s events as we continue to study and think about market structure. So, thank you and welcome.

Chairman Mary Schapiro:
Robert.

Director Robert Cook:
Thank you and good morning. I am joined today by some of my colleagues on the staff of Trading and Markets, I’ll introduce briefly: Jamie Brigagliano, deputy director; David Shillman, associate director; and Daniel Gray, senior special counsel.

We would like to begin by introducing our distinguished panelists, who are seated in alphabetical order, very briefly: Ian Domowitz, who is the managing director at Investment Technology Group; Charles Jones, who is a professor of finance and economics at the Columbia Business School at Columbia University; Chris Nagy, managing director of Order Routing Strategy at TD Ameritrade; Joseph Ratterman, is the president and chief executive officer of the BATS Exchange; Richard Rosenblatt, chief executive officer of Rosenblatt Securities; Stephen Sachs, director of Trading at Diamond Hill Investments; Timothy Sargent is the chief executive officer at Quantitative Services Group; and finally, Gus Sauter, chief investment officer with the Vanguard Group. On behalf of the staff, we would like to extend our warm welcome and thanks to each of you for participating in the roundtable today. And we are very grateful for your willingness to take time out of your busy schedules today to share with us your views and perspectives. And we are very much looking forward to learning from the insights that you can share with us.

As Chairman Schapiro noted, the framework for our roundtables today was really established by the concept release, and each of the roundtables will address a key issue raised in the release. This roundtable will discuss the overall market structure and performance, the second, high frequency trading, and the third will discuss undisplayed liquidity. The Commission has heard a variety of concerns about particular aspects of the current market structure, but it has also heard the view that recent improvements to the equity markets have benefited, both individual and institutional investors. So, one of the main goals of the concept release was to assess more broadly the performance of the current market structure. The release specifically requested comment on the performance from the perspective of long term investors, noting that long term investors are the market participants who provide capital investment and are willing to accept the risk of ownership in listed companies for an extended period of time. But the release also solicits comments on whether it is practical to distinguish between the interests of long terms investors and short term professional traders when assessing market structure issues. More generally, the release solicits comment on what metrics should be used to measure market structure performance over time and how the markets have performed in light of those metrics.

So these are the issues that we very much look forward to hearing from you all during this panel. Our agenda has two parts, first we will hear brief opening statements by each panelist, and then second, we will have questions for our panel by commissioners and the staff in what we hope will be a highly interactive and informative dialogue. Each panelist, as well as any members of the public, will also have an opportunity to submit written comments on any of the issues being discussed today until June 23 through the SEC’s website at sec.gov. Because we have a lot of territory to cover in a relatively short period of time, we ask that the panelists limit their opening remarks to approximately three minutes. So, we will, without further ado, proceed. And, Ian, would you like to start us off?

Ian Domowitz:
Thank you very much. And Chairman Schapiro, thank you and the rest of the commissioners for inviting me here today. I would like to — and I actually see several faces out in the audience who know my penchant for history, so I would actually like to begin with a short discussion of the events of May 28. You may have thought it was May 6, but it is May 28, 1962. At that time, IBM, which was the bellwether, virtually the index of the time, fell by almost 5 percent in less than 19 minutes. Smaller stocks fell between nine and 10 percent in a period of less than 15 minutes. This was a case that was investigated, let us say, quite thoroughly by the SEC at the time. And it was the original flash crash. Amongst other things, it was concluded at that time that the specialist system, which was responsible for maintaining fair and orderly markets and providing liquidity to the market, actually did fail. This wasn’t just a question like in 1987 of the tape running late. I mean, and it did by between 2 and three hours if memory serves. But no one stepped in front of the freight train.

So that is relevant for the events of May 6 because, as I look at it, May 6 actually was indeed a liquidity crisis, but not necessarily an issue with respect to market structure per se, depending upon how broadly you want to paint the picture of what market structure is. I mean, ETFs as a class are a great example. I mean, these have been identified in the latest report from the Commission staff as being the most affected of classes of securities on May 6th for the very simple reason that 70 percent of all trade breaks; in other words, all securities that fell more than 60 percent in that brief period actually were ETFs. It is certainly true that price discovery failed, all right. That evidence is incontrovertible and, frankly, you don’t need fancy statistical techniques. You don’t even need the newspapers to tell you that. A simple graph would do it. What may be less understood is that it actually was a liquidity failure, that if one looks at the various statistics around trading activity for ETFs, they were not dissimilar to that of other securities.

On the other hand, liquidity provision, if you like, was quite different, especially relevant to the underlying baskets that the ETFs are intended to track. Liquidity literally fell to the floor. When I say to the floor, I mean to virtually zero, to levels that were 90 percent down or less, in some cases literally to zero over the course of the event. And that correlates almost exactly with the detachment of the prices from the basket, all right, over time. And that was true even for the most liquid ETFs, even for ETFs that did not fall 60 percent, something like the SPI, all right, the S&P 500, where it is certainly true that the ETF fell over 10 percent and the actual index fell 8 . There was indeed a disconnect. But this was a question of liquidity provision. In other words, it is demonstrable, at least from my desk, that it was liquidity in terms of even the depth of the aggregated order book in the U.S. that actually dried up.

So, in a sense, although I think I will have several things to say about price volatility and performance of markets as we go along, I’d like to conclude with a very basic statement. And that is: liquidity is provided by market participants; liquidity is provided by investors; liquidity is not provided by exchanges; it is not provided by alternative trading systems. This is a more behavioral issue. It is certainly true that market structure today may make it easier to withdraw. It also makes it easier to add. Speed in and of itself, I don’t believe, should be the defining characteristic of debates over market structure. In any case, as I say, this was a liquidity event, at least from where I sit, and liquidity is both a responsibility and a privilege of the investor, whether that be the individual investor, the proprietary trader, the so-called high frequency market maker or, frankly, the market makers that actually do have responsibilities, that I think we’d all agree were not unnecessarily unmet, but back to 1962, at some point, the freight train actually runs very quickly and people do step out of the way.

Thank you.

Director Robert Cook:
Thank you, Ian. Charles?

Charles Jones:
Thank you, Robert, and thank you to the commissioners and thanks for the opportunity to participate in this morning’s panel. I appreciate the focus here on short term volatility, which I will define for our purposes here as price moves that are quickly reversed, because I think it is absolutely an important market quality metric, and it has always been an important metric. And it was indeed mentioned in the Commission’s concept release earlier this year.

The events of May 6th, I think, have focused our attention on short term volatility, I think, because unlike some of the other inside baseball measures that we sometimes use here, short term volatility is pretty easy for the public to understand and observe, and it is pretty unpalatable. Continuous highly automated markets I think have lots of advantages, but if there is an Achilles heel to continuous highly automated markets, I think it may be short term volatility in times of market stress. In fact, the academic studies on this suggest that call auctions may be the way to go at times of market stress, that they lead to better price discovery and less short term volatility in and around the auction. In fact, it is not well known in the popular press, although I think a lot of the people in this room are aware of it, but in fact on May 6, what really brought us out of the flash crash was a five-second pause in CME, E-mini S&P 500 futures. That five-minute pause was due to severe order imbalances, and it was that five-second pause that basically marked the bottom of the flash crash. Once there was five seconds for order flow to come into the market, the futures market stabilized and we began to recover from there. So, given those results in the academic literature and admittedly one observation from the flash crash, I think, you know, harmonized circuit breakers that are being considered make a whole lot of sense with some sort of orderly restart.

There are some people who have argued that there might be a magnet effect, that the circuit breakers might draw prices towards them. But my reading of the literature is that people have had a very difficult time finding any evidence of a magnet effect. If there is something there, it is a very minor effect. It is not something that we should be first order concerned about relative to the advantages of sort of calling a time out and restarting things. So, I think that certainly makes a lot of sense. And I think harmonized circuit breakers will also have another side benefit, and I am thinking in particular of the problem of broken trades. In my book, broken trades are really the mark of a broken system. If the market structure is working right, we should really have almost no broken trades. Those broken trades should not happen because both parties to the trade need to have some confidence that the terms of those trades are going to be honored at the end of the day.

Finally, I applaud the Commission in its efforts to get better trade and audit trail data in order to beef up its ability to quickly analyze these kinds of events. I would also encourage the Commission to really to add human capital and other resources to this whole effort. I think it’s not just about the Division of Trading and Markets; it is also about enforcement. It is about all of the roles that the SEC performs. So, I would encourage you to continue on that path.

Thank you.

Director Robert Cook:
Thank you. Chris?

Christopher Nagy:
Chairman Schapiro, commissioners, division staff, thanks for the opportunity to participate on this panel on market structure today. I am Chris Nagy, managing director of Routing Strategy for TD Ameritrade. My firm, TD Ameritrade, based in Omaha, Nebraska, was founded in 1975 and was one of the first firms to offer negotiated commissions to individual investors following the passage of the 1975 amendment. Over the course of the next three decades, TD Ameritrade pioneered technological changes such as touch tone trading, and then Internet investing, to make market access more accessible and affordable by individual investors.

TD Ameritrade has long advocated for market structures that create transparency, promote competition, and reduce trading costs for individual investors. As technology rapidly advances, it is evermore important that the SEC complete the comprehensive review that it is undertaking to insure that the U.S. markets remain the most competitive in the world. On behalf of TD Ameritrade’s 7 million clients, I submit the views of our individual investors. TD Ameritrade believes that it is wrong to spend too much time focusing on the difference between long term investors and other investors. The simple fact is that when an active trader and a long term investor submit a trade, they both demand three things. First, they want the order filled quickly, two, at the price they sought or better, and three, for the entire size of their order. It is noted in our comment letter in the concept release, TD Ameritrade believes that the Commission can improve the quality of the markets by, one, incentivizing the markets to provide size enhancements; and two, increment reform, insuring that both coding and trading are conducted no lower than a penny. The most important steps that the Commission can take would be to incentivize market centers to stay in the market, maintain two-sided quotes, and, most importantly, to post size.

The May 6 market events demonstrated that today’s market contains many players who use their liquidity opportunistically. In times of market duress, liquidity cannot be so fleeting that it ultimately scares off retail investors from the markets. Increment reform is another important piece. Increment reform means that the Commission takes steps of limiting all coding and trading to one cent minimum price variations. TD Ameritrade believes that this step will go far in protecting retail resting limit orders and thereby instill investor confidence. In response to the May 6 market event, TD Ameritrade supports the Commission’s efforts to establish market-wide circuit breakers. We know however that, to be truly market-wide, the circuit breakers should mirror each other. After reading the 10 proposals by the different market centers, they differ from each other in small but significant ways. The Commission should fix this.

Furthermore, the results have been a good deal of debate regarding the causes of the May 6th market events. Some have alleged that retail market orders and stop orders contributed to the downturn. I can tell you, from the TD Ameritrade perspective, market orders and stop orders are important to our clients and, looking at our own data, we do not believe that there is any basis that these orders made things worse. In fact, TD Ameritrade’s client market orders and stop orders were within average daily volume parameters on a percentage basis. Prohibiting market orders and stop orders would be a significant overreaction and would wrongly deny freedoms that investors enjoy today.

As for whether today’s markets are fair, well, in certain aspects, no. When markets are supposed to open at 9:30 and don’t, when liquidity is available one second and gone the next, or when the best quote is a stub quote, one must question the fairness. However, when focusing on dark pools, high frequency trading, and co-location, as it has been doing, the Commission is taking the right steps to insure the markets are fair to all investors.

Thank you.

Director Robert Cook:
Thanks, Chris. Joe?

Joseph Ratterman:
I would like to thank the Commission and the staff for providing a public forum to discuss these important topics and providing BATS Exchange an opportunity to participate in these discussions.

Many of the topics under consideration in the concept release and under discussion today reflect on past policy decisions made by the Commission, such as the order handling rules, Regulation ATS and Regulation NMS. Collectively, these initiatives have injected competition into the market place. Consequently, the cost of trade execution has dramatically declined. Recent academic studies have verified that transaction fees have fallen and trading spreads have narrowed in the last several years, which we believe has led to a better trading experience for all market participants.

While we believe investors have realized substantial benefits from the current state of competition in the markets, the events of May 6th serve as a reminder that we must remain vigilant about evolving and optimizing our market structure on a continuous basis. I have been proud of how our industry has mobilized quickly to address the particular structural gaps highlighted by market events that day. The exchanges and FINRA and the Commission staff have been working cooperatively to implement several solutions, some of those short term, some of those long term, to prevent a recurrence of these events. Longer term, BATS supports the introduction of limit up, limit down functionality which would prevent clearly erroneous transactions in the first place. This type of approach would allow trading and price discovery to continue within known thresholds while liquidity has an opportunity to come back into the market. The concept really frames the discussion of fairness of today’s equity markets by focusing on a variety of market structure issues, including issues related to high frequency trading, whether particular trading strategies have emerged that are detrimental to the industry, and the impact of co-location and direct market access fees.

Nearly all equity trading in the U.S. today is automated in some fashion and can exhibit characteristics that fall under the umbrella label of high frequency trading. These characteristics include direct access to a market, the sending of a large number of orders into the market, orders generated by computer algorithms, trading through a co-located broker, or subscribing to an exchange’s direct data feed. Future regulations targeted in this area should take into account that the phrase “high frequency” may more broadly describe the state of our market than it does any particular segment of trading participants. In addition, I don’t believe that efforts to judge the benefits or detriments of particular trading strategies are useful. Absent fraud, the diversity of trading strategies in our market has produced a healthy, vibrant, and sustainable ecosystem that allows traders and investors with a wide range of investment needs and differing time horizons to find each other in a network of well-connected liquidity pools.

Lastly, with respect to co-location and direct market data feeds, the latency improvements experienced by firms who co locate can dramatically reduce their trading risk enabling them to more efficiently deploy capital which results in narrower spreads and enhances decreased trading costs for all investors. BATS believes that so long as such services are provided on a fair and transparent basis, the benefits enduring to all investors are real, intense competition pushing securities to their fair value, narrowing spreads and fast executions against firm and automatically assessable quotes.

Thank you, and I look forward to participating on today’s panel.

Director Robert Cook:
Thank you, Joe. Richard?

Richard Rosenblatt:
Thank you very much for the opportunity to participate this morning. I have been an agent for over 40 years. My firm does not have its own trading account and never has. Our goal is simple, to give our institutional customers a competitive edge when investing their clients’ money. We do that by affecting transactions on their behalf, analyzing the trading process, and writing extensively on world markets and how to use them. We have long appreciated the benefits of automation and efficiency in our markets. And actually, in the late 1980’s we were the first firm to offer electronic direct access to public institutional investors. Like everyone in the industry, we are self interested, but unlike most, we are not conflicted in our obsession to make our customers look good. Accordingly, we have a keen interest in markets being as efficient and effective as possible.

So are they? For the most part, yes. The market structure transformation of the last 15 years has brought investors many benefits. Decimalization and automation have dramatically narrowed spreads while enhancing liquidity. Regulation has fostered competition between market centers and driven down the fees brokers pay. But this transformation has come at significant prices. Exchanges, for instance, changed from being not-for-profit utilities to public companies that seek to maximize per share earnings. This places a greater burden on our regulators to ensure that the investing public and, by extension, our nation are well served. Competition between markets reduces costs to brokers, but only competition between orders improves price discovery. Our fragmented markets, aided by such practices as payment fraud or flow, an internalization of damage pricing efficiency. As more and more market venues spring up to serve specialized order flow, institutional blocks, retail market orders and algorithmic trial orders, investors with different investment goals increasingly are unable to access each other’s liquidity. The result: brokers, always required to seek the best possible execution, are today held to a standard that is impossible to measure.

Then there was the so-called flash crash of May 6. It was embarrassing. It shook confidence in our markets, and it was avoidable. Exchanges cancelled hundreds of what they deemed to be clearly erroneous trades from that afternoon. The errors, however, were not in the trades but in the market centers that executed them. Fragmented high speed markets without a mechanism to slow them down during times of stress, cannot discover the right price, resulting in trades executed at absurd prices. In other words, the market structure like this guaranteed that a flash crash would happen. The only question was when. I understand why some would try to shift the spotlight to the use of market orders or even blaming NYSE Classic, the only market that worked that afternoon, but the clear culprit was the commitment to high speed, whether it made sense or not.

I would like to wrap up my opening statement by offering some additional guidelines that I believe should govern market structure. Published quotes should always be rewarded with priority over reserve and hidden liquidity; disclosure of where brokers route orders and why should be expanded; and finally, on the subject of whether the market structure is fair for longer term investors, long term and short term investors provide liquidity to one another, and attempting to define time horizons and tilt regulation toward one end of the spectrum or another is not a good idea. Grading U.S. markets today, I would give us a solid B; nothing to be ashamed of, in fact the best in the world, but not good enough. I am happy to elaborate on any of these views and answer questions you may have.

Thank you again for inviting me to be here this morning. It is an honor.

Director Robert Cook:
Thank you very much. Stephen?

Stephen Sachs:
Good morning. Thank you, Chairman Schapiro and all the commissioners, directors, and staff, for inviting me here to speak this morning on these important topics. The views that I will share today are my own personal views and not necessarily those views reflecting my firm, Diamond Hill Investments.

I will start by saying that I strongly support the Commission’s examination of the current market structure and appreciate the opportunity to provide my views from the mutual fund and asset management industry perspective. As you know, the vast majority of Americans get their equity exposure via some sort of managed product such as mutual funds, ETFs, or managed accounts. I think it is extremely important that you hear from all interested parties on these topics, but I think it is imperative that you hear the voices of those representing the vast majority of the investing public in this country. To that end, I also encourage you to continue to engage the Investment Company Institute and to continue to seek all of their input on these topics. The ICI does an excellent job of reflecting the views of its membership.

The concept release on equity market structure release earlier this year was a encouraging sign to me as it relates to the evolution of market structure here in the U.S. While we have seen extraordinary innovation and evolution of our market structure over the past decade, it has led to extreme fragmentation, varying rules amongst exchanges, and alternative trading systems. While I think competition is good and is a major driver of innovation, we have clearly reached a point where this fragmentation is creating significant issues. Not only did the market action of May 6th demonstrate this, but I can tell you from a day to day trading perspective that fragmentation in the markets is an issue. While all of this competition and innovation has led to markets that can and does handle exponentially more trading volume then we did 20 years ago, we have created an environment that is far more difficult to navigate. If this relates to trading costs, while my explicit costs have come down over the past two decades, I am not convinced that my implicit trading costs are better by an appreciable margin. While bid/ask spreads have tightened considerably, the depth of book, resting liquidity, and average trade size have all decreased substantially in our markets.

I support a multi-exchange, multi-trading venue structure for our equity markets. As I have said, I think the competition, and not the innovation that comes from this structure, has more advantages than disadvantages. However, I do believe that the barriers entry into this business should be high, and the regulatory responsibility should be equally as high and shared by exchanges and trading venues alike. This, I believe, would reign in some of the fragmentation that we face.

The other main issues we are facing and those outlined in the concept release — high frequency trading, dark pools, and order routing practices — are all significant issues and have ignited some very contentious debate. I am not opposed to high frequency traders operating in the market place. I am not opposed to these firms acting as electronic market makers or executing arbitraged strategies. I do believe that there should be an obligation imposed upon these firms to act as true market makers, and these obligations should be consistent across exchanges and venues. I am, however, very opposed to high frequency trading strategies such as order anticipation and momentum ignition. These strategies only serve to detect and profit from institutional order flow at the expense of that order flow and thus the long term investor represented. I urge the Commission to examine these facets very closely and take appropriate action. I also urge the Commission to examine the order handling and routing practices of broker-dealers. There is very little transparency into where and why brokers route order flow and almost no way for institutions or retail investors to police these practices on our own. We need to look at the regulatory requirements in this arena and help ensure that the conflict of interests are eliminated or at least minimized.

On the topic of dark pools or undisplayed liquidity, I will tell you that these pools are very valuable to the institutional investors. While I prefer to have all this liquidity displayed in the market and accessible, given the current market structure, this does not seem probable. I support the disclosure of volume traded in these pools at the security level and on an aggregated daily basis. I believe this would help lend the right transparency to this undisplayed liquidity.

Ultimately, the question of how well our current market structure is performing is the right question to ask, and it is the heart of all the issues that I have mentioned. And these issues are the direct result of the market structure changes we have made over the past decade. Unfortunately, I think the answer will be much more difficult to come by.

Again, thank you for having me here today.

Director Robert Cook:
Thank you. Timothy?

Timothy Sargent:
Good morning. Hi, I am Tim Sargent. And I think that the reason I am here to make comments today is that my institution that I represent, Quantitative Services Group, is in the business of uncovering and examining best execution, especially from the perspective of the institutional order. And one of the things that, in summarizing our work in this space, I think that the market structure in itself lends itself to the real challenges of the institutional investor getting the size of liquidity that they are looking for. I think that the biggest issue that we confront in our analysis activities is the unintended consequences of anonymity. From our standpoint, being able to examine the details of execution quality come down to those individual trial order executions that are now required for institutional order flow to be conducted in these markets. As has been previously referenced, these smaller actual average execution size and the necessity of institutional orders to be cut into increasingly smaller pieces poses a lot of different threats to the cost equation for these clients. So, our analytics and the research that I think bought me the podium today tend to operate in a corner of all institutional activity.

And I think that, if there is one thing that I think needs to be highlighted, that the average experience is not where a lot of the issues lie, and for institutional orders themselves, our institutions are really concentrating on those very expensive trades. And decimalization and the impact of the structure of markets has had a real impact on that. It has already been highlighted that the implicit costs are very difficult to untangle. And as we go back through and use a very forensic technique of looking at that cumulative price concession that institutional orders have to eventually pay to get their positions done, we do find some anomalies in that area. And the most pressing anomaly that we find, in this anonymity in this world of the ability of pattern recognition, small spreads and other sorts of challenges that benefit a certain part of the trading community, we routinely find that large footprints are created unnecessarily for many of our clients, and it is our hypothesis that, of course, those large footprints are the results of different motivated traders who may recognize the order flow as it enters the market. And I don’t think that it’s an extension of belief to think that these profits occur from some place. So, from our standpoint, we are not concerned about indicting any particular practitioner in the market place; our role is really to alert that this is occurring, that our clients have mechanisms to be able to manage this order flow, and the evolution of those markets are essentially that we have to be able to provide these techniques to the consuming public so that they can address a very competitive market place.

But from our standpoint, this market place is the same competitive market place that was in place with, you know, the traditional specialist model. People — most institutional clients are aware that there are costs to trading, and the areas where the anonymity starts to break down is some of the natural governing principals, especially when I entered the business and you had somebody upstairs negotiating and trying to pull together natural liquidity from an institutional standpoint or you had the specialist on the floor trying to keep orderly markets — many of those things are diluted by, you know, by the electronic, you know, environment itself. And so, from our standpoint, what we have had to advise many clients to do is essential adopt the same trading mentality as their competition, which requires them to understand and vary their patterns of behavior to essentially protect themselves from the marketplace. And so, from, you know, this illusion of liquidity — and I think that it has been stated pretty clearly for the public — isthat the illusion of liquidity rests on the modus of the market participants. And so, to the extent — I think it was well said recently, that you are going to get liquidity on the bid and offer side for many of these participants until there’s a signal. And I think we saw the biggest signal on the 6th, where people run and then liquidity is pulled away. And so I believe that some of the things that we can suggest that would be beneficial would certainly be the ability to be able to look at the quality of execution by venue. And so, if there is a recommendation right now, most, venues themselves don’t report directly to the tape. So, for the consolidated tape that every investor would have access to after the market closed, it would be valuable to add additional information sets there to promote the analysis.

Thank you.

Director Robert Cook:
Thank you. Gus?

Gus Sauter:
I would like to thank Chairman Schapiro and members of the Commission as well as the staff of the Division of Trading and Markets for giving me the opportunity to speak here today and to participate. I am pleased to represent the Vanguard Group and our more than 10 million investors. We have two core beliefs at Vanguard with respect to successful investing. The first is that investors should focus on long term investing. Focusing on short term investing, in our view, leads to a pretty bad recipe for most investors. The second belief we have is that is it very important to minimize the cost of investing. And I will focus primarily on transaction costs.

I would say that we believe that our investors have benefited very significantly over the last 10 or 15 years from all the structural changes in the market place; the changes to the order handling rules, Reg NMS, decimalization, improved technology, the advent of high frequency traders. It is impossible for me to disentangle all of those effects over the last decade, but we do note that our transaction costs have been cut by more than 50 percent in that period of time. With a decline in that amount, compounding those savings over time, it leads to very, very significant and a different outcome for investors if saving for retirement over a 30 year time frame. You will have as much as 30 percent more compounding your returns with half the transaction costs that you would have otherwise. So it is very meaningful to have lower transaction costs.

I would say that we don’t own an alternative trading system, but we do use them. We are not high frequency traders, but we do believe that high frequency traders provide benefits to the market place, that they knit the very fragmented market place together. We have now more than 40 different exchanges or venues on which to trade, and it is important that they are all acting in concert, and we believe that high frequency trading accomplishes that. It leads to market efficiency and price discovery and, ultimately, reduced transaction costs.

There is a significant flaw in the market structure, though, and certainly May 6th revealed that. I would like to support the Commission in their consideration of two things. First is the displayed limit orders, because they are the building blocks of transparent price discovery. And the second would be the current discussion about single stock and ETF circuit breakers. I will turn to the issue of limit orders first. There’s been a lot of discussion about the competition among exchanges, and clearly Reg N and AHS and the change in the order handling rules were promoted in order to increase competition across exchanges. However, as stated earlier, we do believe that it’s more important to promote competition among orders. That is truly where price discovery occurs.

There are three things to consider if we want to increase limit orders and provide greater liquidity in the marketplace. First would be depth-of-book protection. The current rules protect the top of the book but not the depth of the book. We also support the Trade-At rule, and then finally a third consideration we’d like to propose is investor education.

With respect to depth-of-book protection and the Trade-At rule, we believe that both of those provisions would provide for greater liquidity, that there would be an incentive for traders to enter limit orders. Today, there’s really very little incentive to enter limit orders when you can participate without having to display your interests. We think it’s very important to educate investors. Very few realize the very significant risks they were facing placing market orders and stop-loss orders, so we think they need to fully understand that. We do believe that it’s possible to, in effect, promote a market order even requiring the sole use of limit orders. So, in other words, if there were a rule that said you could only place limit orders, you could still initiate a market order by saying, “I’m willing to pay a thousand dollars a share for a given stock” or “I’m willing to sell something at one cent a share.” That, in effect, would work the same way a market order would work. The advantage is that’s an intentional statement of where you’re willing to trade as opposed to one gathered by default. I would note that individual investors probably received the brunt of the blow on May 6, as their orders probably were the ones that were right at the limits that were not cancelled because of their market orders dragging them down.

Turning to circuit breakers, the final thing I’d like to discuss, we think it is important to require a timeout to let investors become calm in times of stress. I’d say that the New York Stock Exchange accomplishes this through their slow-down mode. In fact, we know that there were no orders that were cancelled on the New York Stock Exchange on May 6. Of course, there were very few orders traded during this stressful time period, so I think actually the slow-down mode was very effective at, in effect, providing a circuit breaker. We think we need a similar type of function in the electronic markets. The electronic markets were free to trade with whatever liquidity remained on those markets, and we think that the single stock circuit breaker or single ETF circuit breaker would provide an automated way of the same service that the New York Stock Exchange has when it goes into slow-down mode. Such a rule would give investors a time to settle down and allow for liquidity to be replenished. In fact, if you knew that you could trade an ETF that declined down to one cent, given five minutes to think about it, wouldn’t you have purchased that down only 20 percent? I think most of us would have stepped up and provided liquidity if we had time to think about the fact that the world was not ending, it was just the unfortunate circumstance that everything unfolded so quickly that there was no liquidity left on the book.

We think this is a far better solution than applying the Clearly Erroneous Trade rule. The Erroneous Trade rule, in fact, disincents liquidity. If you were a market maker or high-frequency trader and you’re subject to an erroneous trade, you’re going to pull back and stop providing liquidity because you’d be concerned that by providing liquidity, that could be a trade that is cancelled and your hedge on the other side might not be cancelled so you’re going to step back. And we think that’s what really happened on May 6, that everybody was afraid to provide liquidity because historically the Clearly Erroneous Trade rule is applied when the markets are off 10 or 20 percent or individual trades are off 10 or 20 percent. So if the markets are approaching that level, those that would have provided liquidity at, say, down 10 percent would step back because of fear of having their trade cancelled.

In summary, I’d like to say that we do think that there have been tremendous improvements in market structure over the last 10 or 15 years. We think that those benefits should be preserved. We think that there is certainly a problem. Certainly May 6 told us that. We think that there can be changes to the market structure that preserves the benefits that the market structure currently provides. And I’d say we look forward to working with the Commission on this very important effort. Thank you.

Director Robert Cook:
Thank you, Gus. So that concludes our initial opening statements. We’ll now move to a question-and-answer session with questions coming from the commissioners and from the staff. And we appreciate all of you limiting your remarks as you did, and we know you have a lot more to say and we look forward to hearing it. So we’ll start with the questions.

Chairman Mary Schapiro:
Thanks, Robert, and I’m not going to make my colleagues go in our traditional order so any time you have a question, please feel free to jump in, but I’ll take the prerogative of the chair and start. I have sort of a broad question or comment and then a very specific one.

As I listened to all of you, and I thought your comments were enormously helpful to us, I heard benefits of our current market structure pretty clearly, and I heard concerns with our market structure, fragmentation, and severe and extreme fragmentation were two adjectives I actually heard used, being the primary one. And as I went through a number of the comment letters, and I can’t tell you that I read them all at this point, I see the benefits and the problems with our market structure sort of shaking out this way. The benefits are that volumes have increased tremendously. The markets can handle enormous volumes compared to 10 or 15 years ago. Spreads have narrowed. Costs have declined, depending on how you define them. Execution speeds have increased, all to the benefit of investors long-term and short-term, and I take the warning about trying not to define those quite so specifically perhaps.

On the other hand, high cancellation rates have resulted in flickering quotes, volatility’s increased, the size of executions has declined pretty significantly. There’s an illusion of liquidity in that it looks like we have lots and lots of liquidity because we have lots and lots of volume until we don’t have liquidity, which can happen in a nanosecond. And the public perception of the markets is that they’re not particularly fair.

So, any thoughts on how the Commission weighs these two sides of the equation in thinking about how we prioritize what we should do to improve the quality of our markets? And I really open that up to anyone, although you look interested, Mr. Rosenblatt, in responding.

Richard Rosenblatt:
Thank you. I’m always interested. Markets do not assign value to equities, the investing public does. Markets discover price and when they’re doing their job well, that price that they discover is very close to the valuation that the public would place on an equity. When we have disparity between those two prices, it is the fault of the market structure that it has been unable to approximate the value that investors place on an equity. We saw it on May 6. My hat’s off to Ian. He came up with an example that actually pre-dates my career.

[laughter]

Very hard to do. I was here for the crash in ’87, and I have seen many other minor dislocations in the ability of a market to accurately append price. That said, when an investor, as a community, decides that a $40 stock is worth $20, it is not the job of the marketplace to say, oh, no, we’ll stop it at 35. It is the job of the marketplace to get that equity to $20 in a reasonable way, because the marketplace itself is agnostic. It doesn’t know what the value that the public wishes to place on the equity is.

So, a lot of us have suggested that when there are dramatic moves the market should slow down. Allow the public — when I say slow down, I’m talking about trading. You never slow down information dissemination, because that is what fuels the marketplace. In extreme circumstances, you may want to stop and go to a cold market. Most marketplaces have that ability. I think these are very important distinctions. I think automated markets have added a lot to the landscape. However, they’ve taken away people, and computers can’t judge reasonableness. So we do have the potential in automated markets to see absurd results.

Chairman Mary Schapiro:
A public company sitting here, though, might say to you high frequency traders, using that term very loosely, doesn’t really care whether my stock is worth $40 or $20. It’s not about driving to the price that reflects the economic value of a share of stock in my company. It’s about lots of other strategies and motivations and that that’s hurting the capital formation process, and that’s what’s driving our markets, not a genuine sort of supply and demand give and take based on real economics.

Richard Rosenblatt:
I think to some degree that’s a valid comment. I think our markets are so fragmented that even with rapid information dissemination the ability of an investor who would be very willing to invest in a company at a higher price than it might trade at does not have access to that marketplace. And by the time they receive the information, it’s too late for them to take action. I think it’s a big challenge that we face. I think there are benefits to competitive markets. Fragmentation has been the biggest detriment that we’ve had to face in terms of accurately reflecting public values for equities. I guess I should give somebody else a chance.

Charles Jones:
Can I jump in on that? I think it’s very hard for us to look at the data and say, you know, “Should IBM be at $80 or $120?” That’s not something that’s going to be easily discerned in any sort of empirical study. But what does seem to be evident in the data is that the current market structure seems to give us, most of the time, quite efficient prices. The short-term volatility, these price reversals, most of the time are not very big, and what we seem to have found from the current market structure is that this competition between liquidity providers and the technology that we have is basically giving us, as best we can measure, prices that seem to be efficient. So I would say that, to the extent that we can measure this, that the current structure, by and large, most of the time has been very beneficial for price efficiency.

Ian Domowitz:
I suppose that one of my themes today will be to try to make the point that we shouldn’t necessarily confuse issues surrounding market structure with issues surrounding behavior as we have seen them before. So let me give you a couple of examples. The first relates directly to your previous points, all right?

The issue with respect to speculators, all right, in the futures and options markets is quite old, pre-dates electronic market structure by a great deal of time. And it certainly has been conceded, at least I believe so, that the benefits to the hedging community of activities of proprietary trading shops in Chicago have been largely for the good, all right? I mean there have certainly been exceptions historically. But that is an old debate and should not be confused with the notion of electronics.

Another might have to do with execution-size declines. Now we are tempted to ascribe that also to market structure simply because the timing is in some sense fortuitous, because it more or less coincided with decimalization. Now we might stretch the point and say decimalization is a part of market structure. It is certainly one of the rules by which we live. But, once again, one might divide things up. In other words, the fact that depth of the quotes fell sharply, in my personal opinion, was not a question of electronic market structure. It was a question of reducing the minimum price variation, all right, which could have been done in any market. Of course some business models failed, all right, due to that. That was expected. And others might have been if electronic market structure hadn’t been in existence.

So, you know, there are cases in which the market structure actually is pertinent. So I heard the term automated markets have taken away people. Now, I remember very well actually testifying in front of the Senate Banking Committee back in the very early days of electronic market structure with respect to competitive advantages and disadvantages when the honorable senator from Illinois took me to task for the fact that, in her estimation, four out of six people in Chicago would lose their jobs if indeed markets were automated. That is absolutely true. Automated markets have reduced costs to the industry by a phenomenal amount. I mean, it’s hard to give you a real idea in current terms, but I can tell you that people used to estimate that a 60 to 70 percent drop in operating costs was possible by eliminating trading floors and moving to electronic markets. Those costs have been passed along, or I should say those cost savings actually have been passed along. Even the building of such a market, the treasury floor in Chicago cost $187 million to build. At that time you could have put together a fully functioning market that would have handled thousands of securities for $20 million. And some of you may remember the London Stock Exchange abandoning a $400 million project, right, in favor of going electronic. So there were cost savings there at the point of the provider.

Now, one last thing, because I realize I’m taking up some time, but there is one thing that I believe should be refuted and I think Mr. Sauter certainly began. It’s been mentioned that somehow someone doesn’t believe that implicit costs have actually gone down with the advent of electronic trading and electronic market structure. As someone who actually oversees and as part of my world a facility for serving roughly 300 installations with respect to transactions cost analysis and, as you can tell, I have some liking for history, I can tell you that statement is absolutely false, all right? I’m more with Gus. In other words, 50 percent may even be an understatement as a matter of fact. We certainly did observe a blip in so-called implicit transactions costs starting with the Lehman collapse. As a matter of fact, starting as early as August, 2007, coming to volatility, there was an extreme spike in volatility at least at that time. We hadn’t anticipated October of 2008 yet. So they certainly went up and then declined. We’re now below 2004 levels, but that might give you an idea. Transaction costs were sliding in the aggregate and at various individual levels, depending on how you sliced the data, for years. And there are a variety of academic studies as well as industry studies that correlate that with changes in market structure as well as the behavioral changes that have been occasioned by those structural changes. So I think, just for the record, all right, that should be set straight.

Director Robert Cook:
Thanks, Ian. I think Commissioner Paredes had a question.

Commissioner Troy Paredes:
[inaudible]

Christopher Nagy:
Thank you, Commissioner. I wanted to go back to your first comment about the markets operating very efficiently, and I agree with you. I think during normal market hours after 9:45 in the morning on a normal day the markets do operate very efficiently. But I think there are two kind of gaping holes that we need to take a look at, and if, you know, I look back at my retail investors and specifically what I classify as our long-term investors, where do they get into the markets? Where do they invest into the markets? And it’s really at two points in time. And if you think about this realistically, it makes a lot of sense.

So the long-term investor generally works all day, and after they’re done working, they get home, they eat dinner, you know, see their family to bed, and typically that’s when they sit down and take a look at their personal financial situation and make decisions about that over the course of the evening. As a result of that, each and every day, 10 to 15 percent of our orders are executed right at market open, right at the market open, yet we don’t have statistics that measure the market open. We don’t have a market that actually opens at 9:30 each and every day, yet we’re measuring, you know, executions in microseconds throughout the course of the day. So that process needs to be, I think, looked at holistically for the long-term investor.

The other time the long-term investor’s in the market is during times of stress, like May 6, and a good example of that would be the person’s at work and walks over to the water cooler and somebody mentions, “Hey, have you seen, the market’s down 600 points?” Wow, kind of commands your attention. I’ve got to take a look at my investment portfolio, my long-term investments, and fear and other factors come in play to that. We need to ensure that during times of stress that the markets remain robust so that the retail investor doesn’t panic, that remains calm and doesn’t, you know, sell into the market during those times.

So I think those are two kind of key points that I wanted to bring up with respect to that, that I think would be important to look at.

Commissioner Troy Paredes:
Just to go back to something that Professor Jones had said, if I heard correctly, and that is that the current structure has been beneficial to price efficiency. I’m curious if you’d just expand a little bit more in terms of which aspects of the current market structure you think have, in fact, been beneficial when it comes to pricing efficiency and [unintelligible] and if others have either a similar view or a competing view, I’d certainly invite others to chime in as well.

Charles Jones:
So I’m thinking in terms here of short-term price reversals, so our usual definition of short-term volatility, so price moves that are then quickly reversed. And the general evidence is that there’s been a solid decline in that which was reversed during the financial crisis but has since sort of come back in terms of its steady decline. And I would also point to what Ian said in terms of thinking about institutional trading costs, the buy side that employs firms like ITG or, you know, Elkins/McSherry, or some of these other trading cost analytics firms, all of them find very consistent evidence that they’re trading costs have fallen quite dramatically. And so that’s another, I think, very important metric that we should be looking at.

And finally, I think, retail investors are going to be looking at — are typically trading not huge amounts of size and so they are the ones who should be focused on spread measures and some of our sort of traditional market-quality measures. And again those, I think, have shown, you know, steady, continuous, pretty much continuous except when punctuated by certain events from 2007 and 2008, pretty steady improvement.

Commissioner Troy Paredes:
And other particular features of market structures you think are accountable for those improvements or are we simply focusing on the results are but hard to show and argue as to cause and effect?

Charles Jones:
I guess I’ll go back to what Gus said at the beginning. I marked it down because I totally agree. It’s very difficult to disentangle all of these effects. All we can really say is that, you know, we started here and we ended here. And so it’s all coincident with all of these changes but a lot of them are very difficult to pin down completely.

Gus Sauter:
Could I offer a comment there that I think it’s important to aggregate liquidity together. That fragmentation really is a huge detriment. So one of the benefits in the current market structure would be all of the participants that do knit everything back together again. If you have 40 exchanges, and think of all the different combinations of or permutations of 40 exchanges trading against each other, you could have very different prices on the various exchanges, and somebody needs to bring those back together again. We’ve long been a very big proponent of a central limit order book, which I know has not gotten much support over the years, but the advantage of a central limit order book is it does concentrate liquidity in one place, and I think that’s what Reg NMS was trying to do by trying to knit everything back together again by honoring the top of the book. So we do think that is a very important aspect of Reg NMS, is trying to knit that back together again, encouraging participants to really, quite frankly, arbitrage one exchange over another to bring prices together. I mean, you know, that’s what high-frequency traders are doing. It’s hard to label exactly what a high-frequency trader is and they have all different types of strategies, but many high-frequency traders are just observing price discrepancies across two markets and bringing them together, so we think it’s wrong to just label high-frequency traders as abusing the system. We think they’re actually helping the system.

Commissioner Elisse Walter:
Can I follow up on that? You talked about essentially mechanisms for bringing the liquidity that is in the market to the fore and making sure that it’s used. Do we need to go further than that and provide incentives for people to step forward with liquidity and to provide that liquidity and if so, what would you advocate?

Christopher Nagy:
I would absolutely agree with that. I think that’s a very important aspect that’s lacking from our markets today. We’ve moved to an environment where we incentivize the quotation rather than incentivizing a two-sided market, and I believe that the way to restore that is to effectively incentivize the electronic market maker or specialist with either — you don’t just want to say, “You have to make a two-sided market,” because there’s going to be no incentive for them there, so in today’s markets I think that would either be through market data incentives or through other — you know, I know you don’t have the authority to do any tax incentives, but I think that would be an encourageable step to have more capital committed into the marketplace today. What we see a lot of times at the inside quotations are literally fleeting quotes. You know, you see a 100- or 200-share bid offer a lot of times at the inside quote and that is cancelled and replaced many times if you look at the quote. There’s no requirement to maintain a quote for any period of time at the inside market. The retail investor, the individual investor, gets frustrated by that. They see a quote one minute and then the moment that quote’s gone, right? So I think we need to work on incentivizing firms to really step up, commit capital over the duration, and the only way to do that is really through, I think, credits and incentives.

Stephen Sachs:
Yeah, I can tell you that doesn’t only frustrate the retail-investing public and I think that’s a very, very good point that you made, that the incentives in the market structure we currently have incent the quote necessarily and not the depth of book. And personally, I’m also a big fan of the central limit order book or anything that generally will bring that displayed or transparent liquidity back in the market. I mean, obviously being the one who made the comment about the implicit, you know, transaction costs not coming down appreciably, you know, we could probably debate that aspect all day long and ultimately the issue of transaction costs analysis is my transaction cost model is very different from your transaction cost model and that it varies from client to client.

But ultimately, you know, the electronic marketplace, again, is not bad by its very nature. But what do we put in place to incent that depth of book, depth of quote, increasing the trade size back to a more reasonable level and then ultimately changing the psychology of how we approach the marketplace, because, again, you know, much of what we analyze in transaction cost analysis isn’t necessarily market structure but how I react to that market structure, you know, and it’s not an easy question to answer.

But I do think the concept of continuing to pull these fragmented markets together, continue to put in some sort of across-the-board regulatory environment that helps preserve that slowing down, if you will, in times of crises, that’s how you incent people to come back in and set those prices as opposed to the exchange trying to set the price, which, as we’ve discussed, is not the exchange’s job.

Timothy Sargent:
I just, to follow up on Steve’s point there, too, is that as you’re going back through in the advisory function and people are looking for solutions, they’re tending to go toward the order parceling size, sort of reinforcing the small levels of liquidity that we see at the inside bid and offer because they accept it. And you know that evolution of a block mentality or the idea that there’s a valuation level on a trading desk, that’s evolved away. We’re in, you know, previously you might have an under-the-market bid in a particular security that may be well aware on the trading desk, but that mentality has been replaced with this idea of trying to be more concerned about the signaling risk of the way that you’re parceling or distributing your order flow up during the day than really representing that true level of liquidity where the institution would actually transact in size. And so if mortgage structure or whatever, however you get there, you’d have to be able to create alternatives, and I do hear in the background that the institutions themselves — I was just at a trader forum meeting where the buy-side institutions are considering, “Well maybe we should just trade with each other again and allow ourselves to be in that environment where that signaling risk is minimized,” you know. And again, you know, I don’t think that that’s a step forward in being able to leverage the infrastructure that’s been created.

Gus Sauter:
There’s been a lot of discussion about whether or not electronic market makers should be required to step up and make markets in all types of environments. We don’t believe that would be an appropriate response, that essentially you’re requiring people to go bankrupt in an event like May 6. And I’d note that people have said, well, the specialist system worked but in fact the specialists weren’t making markets either. They basically went into a slow-down mode. I’m not faulting them for that. I think it was the appropriate response. But I don’t think requiring people to make markets when things are crazy is the right thing to do. I think in order to incent people to make — to provide limit orders, you remove the disincentives to not providing limit orders. So I’d note that if you provide a limit order, you’re essentially providing a put-option to the marketplace. If somebody does not like their position, they can always put their position to the book. And so you’re taking risks by putting limit orders out there and so that is, in itself, a disincentive to placing limit orders. You need to try to minimize that risk, in other words, give standing to those that are willing to display their orders, honor not just the top of the book but the depth of the book as well. And again, the Trade-At rule we think provides protection.

I’d also note that circuit breakers would reduce risk. You basically need to reduce risks of providing liquidity to the marketplace, and I think circuit breakers would reduce the risks to the electronic market makers.

Stephen Sachs:
I think the key to that is what you touched on, is that you have to disincentivize or take away the incentive of the stepping ahead of the order that takes place currently, because the barrier is so low with any increments, to stepping in front of order flow which then will ultimately change the psychology of that particular order, how it’s handled and thus impact the price of the stock, right. Every transaction has an impact on the price regardless of how big or small. You know, that is at the core of the issue. It changes without disincentivizing or making that hurdle higher to step ahead of those orders, you know, the psychology question comes to play and you end up in this, you know, sort of circle of bad logic on how you handle an order.

Chairman Mary Schapiro:
Could I follow up on that a little bit, because historically market makers have had obligations because they’ve also had privileges and they’ve had benefits whether it’s proximity or other things. It strikes me that if you have the benefit, for example, of co-location, shouldn’t that come with some special obligation to the marketplace to provide liquidity, to be there, to not reach across the market, to not take away liquidity, whatever we could end up defining those obligations to be?

Joseph Ratterman:
I’ll go ahead and take that. In our minds co-location is an optimization available to all participants of the exchanges. While individual retail investors will never co-locate nor would we expect them to, they’re going to go through a broker-dealer that has the ability to co-locate. So we don’t believe that co-location provides a significant time-and-place advantage over any other participant. All participants are able to choose that service from every exchange. The prices and availability are — as long as they’re made fair and transparent — we believe this is simply an optimization that allows for more efficient trading for those who choose to use it but doesn’t provide enough of an advantage over other participants that you should have an obligation that goes along with that.

Stephen Sachs:
But doesn’t it depend on what you’re using that co-location for? Because, again, from a long-term investor’s perspective, you are right, you know, they have access through the broker-dealer, for example, the broker-dealer or agents that we’re using. But again, that co-location, that speed is not necessarily always benefiting long-term investors. It benefits liquidity providers, high-frequency traders, so, again, it depends on what their business model is and what they’re in the marketplace doing. As I mentioned earlier, I personally believe that if you’re going to have advantages posed to you for being a market maker, there should be some obligation, whether it be, you know, regulatory or set by the exchange, however you want to define obligation. There should be some cost to entry into that marketplace, which I think currently the cost to entry is extremely low, you know, for these liquidity providers. The offset to that is that they have the ability to pull out of the market, as we saw on May 6, and not that every single liquidity provider did, but, you know, clearly some of them pulled out of that market. So how do we balance that between, you know, again, what the ultimate purpose of our capital markets is, capital formation, capital distribution, you know, the ability for companies to get the capital they need and distribute that capital to the investing public.

Joseph Ratterman:
Good points. You know, from an exchanges perspective, it’s broker-dealers that are our customers not necessarily the long-term investor or that be institutional retail, you know, that’s your role and Chris’s role to provide services to those individuals. From our perspective, it’s a level playing field of all broker-dealers to providing the best possible services to their customers. And that level playing field is available to every single broker-dealer that wants to connect into an exchange.

Christopher Nagy:
I’d just say that the issue here is that, and it was brought up, that individual investor will never really be able to co-locate, right? That was mentioned and it’s just because they can’t have direct access to the marketplace. So right there that puts them at a disadvantage. And the overall issue with co-location, not that it hasn’t provided benefits to individual investors, it has. It’s made executions faster. But the real issue is when you look at the total pipe of what’s going on, you see so many orders and then following that you see so many cancellations in the marketplace. Those costs, that infrastructure, it’s actually borne by the retail investor. Our market data costs are amongst the highest in the industry. Why is that when there are firms with much less capital that are submitting 5 to 7 percent of industry orders each day that are paying far less costs than we are for that data in the industry?

So I do agree, Commissioner, that there should be some thought process on if you are co-locating that with that you have to undertake some certain privileges as a result because not everyone has direct access to those same features.

Gus Sauter:
If you disallow co-location, though, then the real estate across the street from the exchange just becomes the most expensive real estate in town. So I would agree with Joe’s comments that we’re not members of an exchange but we use brokers that provide us with that access. And I think it basically makes the lowest common denominator that every broker has to co-locate.

Ian Domowitz:
If I could return to the higher point of incentives that was raised by the Commissioner. I mean, I love to talk of incentives in the context of market structure, and I need to say that up front, given what I’m going to say next. I think that it’s a very fine line between regulating and overseeing market structure and regulating and overseeing the industrial organization of the securities industry. At the point that we leave talk about how market structure itself might incent people and start to talk about what are essentially subsidies, at that stage I believe we move into the regulation of the industrial organization of the securities markets. Now that indeed may be the overall goal. In other words, I don’t presume, right, to make that judgment. But I think it’s something that needs to be thought about because there was lots of talk about, well, maybe we can incent depth of the quotes, you know, maybe we can incent two-sided quotes, because, by the way, we tried to incent quotes, and we did, and there were unintended consequences, right? Academics and industry alike adopted the so-called maker-taker model. In other words, you got paid for putting your quote up, right? I mean, this was considered a good thing. This followed an extensive literature on the benefits of limit orders and open limit-order books so let’s pay people to get stuff on the book.

Now, I’m not saying that that wasn’t a bad idea. In fact it evolved as a private sector business model. It did not evolve from regulation per se. On the contrary, in the end from the point of view of regulation, we got into price regulation, right, because we saw that things were getting a little out of control with respect to how these incentives, all right, were taking hold. But that’s in a sense the crux of my warning. I mean, you can go very easily from the notion of market structure right into price regulation. Now, it may be, again, that that is indeed the purview of the regulatory agency and something you might want to get into. But my role here is simply to point out that when you get into this type of discussion, you are crossing that line, and I think we all should realize that we’re no longer talking about market structure. We’re talking about enabling technology and incenting certain people financially to use the technology in a particular way that is basically behavioral and would have happened with or without the technology. Please keep that in mind.

Richard Rosenblatt:
I think we might be missing the point here. The only advantage that one investor has over another is a time advantage. Your brilliant idea only becomes profitable when that idea becomes common knowledge. It’s always been the case. And what technology has done for us is provide, among other things, phenomenal growth in our ability to disseminate information. So the timeframe between that brilliant idea and common knowledge has shrunk dramatically.

Now, I remember the day when I could advertise that I would get you into the marketplace in under a minute, which made me about the fastest guy around and I was very proud that. I don’t think I want to brag about that right now.

[laughter]

But I’m not sure that we want to be penalizing speed. Speed is simply a function of better technology. That’s not to say that we don’t want to create an environment where we create advantages for market makers to put up capital and the price of that advantage is that under times of stress, they have obligations. I think that is a separate discussion than what technology has afforded us as an industry in terms of added speed to market. I don’t think that’s new. I just think the magnitude is awesome and tomorrow it will be faster. So where we’ve seen over the last year market speeds go from milliseconds to microseconds, and it again speaks to flickering quotes. It’s a function of what we can do today. Quotes always moved as quickly as you could modify them. It’s just that we can modify them more quickly today. That creates enormous challenges in market structure because our goal is to protect the investing public. That’s where the capital formation in this country comes from and that’s why our markets, our secondary markets, exist in the first place.

So we can’t forget our purpose, but I think creating an environment of incentives to create obligations for market makers under stress is a great idea. I’m not sure we want to tie it to speed to market. It might tie itself because, properly incented, those firms and individuals who might want to fill that role would want to take advantage of being the fastest to market, so you might accomplish both goals.

I also think, if I may comment on a central limit order book, which has always been appealing, it’s very problematic because one of the problems we had with NMS — NMS knew that we were going to divide our markets and reduce the dominance that the New York Stock Exchange had — ECNs had already reduced the dominance of NASDAQ — and did not expect that we would suffer fragmented markets. It was not the intent. The order protection rules were designed to protect limit orders on the assumption that order flow from any market would be driven to those who displayed their orders. The problem was we assumed that we could mandate transparency. Investors will only display as much of their investment intentions as they feel is safe. It can’t be mandated. And the result was instead of seeing markets that were a thousand up, two thousand up, our markets are a hundred up, meaning that the disadvantages of fragmentation were not offset sufficiently by the order-handling rules.

Timothy Sargent:
Excuse me. Our client base that we’re servicing when we’re concerned about co-location, they’re generally more concerned about the speed advantage as it comes to another inter-related subject, which is cancellations, because what they’re really trying to understand is their quality liquidity on the venue or are they exposing themselves to another opening for that signaling risk. And so, from their standpoint, that’s one of the reasons that we were suggesting that transparency in those execution routes at the end of the day would allow for analysis to look at the quality of liquidity on each one of those venues and allow an informed consumer to decide on where to direct their order flow with some sense of confidence. And so to order cancellation and speed — because I think everybody understands that you could be in a situation where you’re on the off bid and offer site across 40 markets, you get essentially an offer listed on one market and with the speed advantage, you’d have the opportunity to cancel all other offers and maybe move to a more aggressive, you know, price position on the bid side of a market, and I think that that’s the concern that our institutional clients are, is that they are aiding and abetting, you know, a profit motive that they recognize has every right to exist but they’d rather not make that contribution.

Richard Rosenblatt:
I’m not sure I’d say aiding and abetting, but the problem is that we want all liquidity in our markets that we can get. I agree that we need to incent transparent orders, and that doesn’t mean that we want to shun dark liquidity. We need it. Because we’ve tried to mandate that it become transparent. It doesn’t work. We don’t want to drive it away from our markets. I do understand how an individual market center can blend their dark liquidity with their transparent liquidity to effectively discover price. Some do it better than others but they all attempt to do it. I’m not sure how we would be able to do that between markets, and this is one of the major problems with a central limit order book. No quality market trades through its own liquidity, and that means the prices that they discover are not inferior to their displayed liquidity or their hidden liquidity. It’s very important. My concern would be with a central limit order book that practically that would be impossible to accomplish, and therefore my focus has always been on incenting more transparency in our markets and then the benefits of Reg NMS, which I think were always intended, could be more realized by driving order flow to those transparent markets.

Director Robert Cook:
Richard, I want to pick up on a different aspect of transparency. Some of you have talked about order routing, order handling practices and the quality of information available at the open. Could you comment on what types of initiatives we might think about undertaking to update the current order execution quality, order routing handling disclosures that we have or other types of disclosures, post-trade disclosures or similar types of disclosures that would allow investors, individual or institutional investors, a better insight into how their orders are being handled and any types of disclosures along those lines?

Ian Domowitz:
Well, it’s a little bit hard to recommend, you know, policy prescriptions, but I will give you an example that perhaps could be if not mandated at least incented. This is the one time that I’ll mention ITG.

At our firm, any institutional client can receive a complete report card at the end of the day as to exactly where the order got executed, including all dark pools that we access on behalf of our clients and all strategies that are used simply on request. We produce hundreds of these on a daily basis. It is our view that that information is critical to the institutional investor with respect to making informed decisions. Now, I agree, it’s end of the day, it’s not, you know, in the middle of the day; there’s been obviously talk about, you know, when to report dark pool executions, right? There’s the hidden order volume release, if I used the right term there, certainly had something to do with when we would actually report such information as an industry and as an individual operator of a pool. But I would claim that it is within the power of the institutional broker to provide that information today. There our electronic markets help quite a bit. I mean, fragmentation makes it a little tougher, but the technology is definitely there.

So there exist private sector models that are in place. I would suggest simply that one want to figure out a way to incent those a little further.

Stephen Sachs:
I’m sorry. As a client of ITG’s and a big user of their particular product, I’m a big fan of the transparency they provide. But it falls short in that it tells you exactly where your order was executed, which particular dark pool or venue. It’s available on a daily basis, you know, at request, which is great. But what it doesn’t tell you is where the order was routed prior to execution, which particular market or venue that order sat in and why. Ultimately, and it may be a misguided concern but I know it’s shared by, you know, many of my colleagues in the trading community is that the information leakage from the order-routing practices, you know, from the institutional order flow and ultimately where those orders sit and reside prior to execution, again, given the fragmented market structure we have, we’re not necessarily opposed to our orders going to different places, we simply want some transparency as to why they’re going there. Clearly the idea of order internalization, you know, falls into this bucket and that, generally speaking, most orders attempt to get internalized at a broker-dealer first. This is not necessarily a disadvantage or a bad thing; it’s not always necessarily an advantage either. So again, providing some transparency on where and why the orders are being routed to at least allow us the information to understand the practices, what conflicts exist, whether they’re routing to affiliated pools or venues first, disclosed, you know, basically on a weekly or monthly basis. And again ultimately I think required disclosure, not necessarily at just the request of the client, but then we need some ability to be able to police the activity. We currently can tell one of our broker-dealers to not route orders or post orders into a certain venue or market, but we actually have no way of policing that. We don’t know other than the trust that we have with our agents and broker-dealers that they’re actually following our routing practices.

Ian Domowitz:
I would agree with, you know, the criticism, all right, although I would disagree with the issue except to the extent of the trust, because certainly when requested and we provide the information sufficient to actually rule some things out, we certainly honor those provisions. I think that would be a deep, deep repudiation of our own responsibilities as a firm not to do so. I will also agree, by the way, that the issue of information leakage, this may or may not be an electronic phenomenon but we write on this ourselves, and it actually is surprising to me just how much information there is — leakage there is from so-called dark pools. So I would back these statements largely speaking.

Director Robert Cook:
Chris, then Joe.

Joseph Ratterman:
Two quick suggestions. One is on the 605 reports that better granularity in time bucketing could be improved and synchronization of the reconstruction of the markets so that orders are properly categorized. Oftentimes there’s an issue where non-marketable orders are marked as marketable based on the time synchronization issues.

And secondly, with regards to recording executions from dark venues, there’s a lot of debate, and I don’t know where we’re going to weigh in finally about whether that should be absolutely real time or it should be end of the day. But clearly, today, you can understand where orders are executing in the displayed markets, but until Rosenblatt’s Dark Liquidity Tracker comes out a month later, you really don’t have an idea where things were happening in the dark markets. We’d like to see that on a much more timely basis, maybe even in real time, with some exceptions for actual, large executions to protect some of that interest. But still, that’s valuable price discovery information that’s missing from the market today.

Christopher Nagy:
A couple of suggestions as it comes to the Rule 605 and 606 Reports. First off, I will say that it’s our belief that they’ve been tremendously beneficial to the individual investor. Although they don’t use them directly, they are the recipient of the quality that broker-dealers utilize in assessing the regular and rigorous review of execution quality in the marketplace. However, as the markets have changed, so too have the metrics, and knowing that those reports now are nearly 10 years old, they are in need of updating.

Some of the things, obviously, in addition to the speed factors, are taking into account some of the new-order types in the marketplace; IOIs and IOC orders, for example, get bundled in to some of the reports and make it difficult to really ascertain the true quality of execution during stressful conditions, for example, or even normal conditions.

In addition to that, and I will say that this is also — which I’m still perplexed by; it’s also in the Circuit Breaker — Stock-by-Stock Circuit Breaker Rules — why are we beginning at 9:45, you know? When Reg NMS was adopted in 2005, there was a little footnote down there and it said that the Commission is hopeful that the industry will resolve the opening on its own and we won’t need to do any 605 statistics. That hasn’t happened at all yet, and I think it’s critical that we get those measures in place, because as I indicated earlier, the individual investor does invest at market opening quite a bit, so I think that’s important.

Another point — and I’ve heard this, you know, rebuffed here quite a bit — are the implicit execution costs in the marketplace. You know, we really don’t know what that aggregate number is or what those are, so there should be a measure for execution costs in there, the implicit execution costs, the shipping charges that we pay for order execution in the marketplace.

And finally, ironically, there’s not a factor in the statistics at all for any sort of size discussion, or any sort of size monitoring at all. You know, the 605 Reports have been tremendously beneficial in promoting price improvement, as the Commission wanted, although we’ve got some questionability as to has that been lowered to the least common denominator in the marketplace? So, we think that could be beneficial.

In addition, I’ll add one more point, and that is broker-dealers like ours and others, we actually publish some statistics on our website to attempt to make it user-friendly for individual investors and how that quality is assessed in the marketplace.

There should be some thought on providing a mandate for broker-dealers that execute orders to posting information, even if they don’t execute, even if they ship those orders off on an agency basis. The technology is available today to be able to publish information, similar to how we do 606 reports. That, actually, would then become useful to individual investors, is they would be able to go right there and see the data rather than in the pipe-delimited format that’s required today.

Finally, the Rule 606 Reports could clearly utilize some expansion on that and should apply more broadly than they do today and perhaps a lot of the factors with that should be clarified and improved.

Director Robert Cook:
Thank you. Commissioner Casey?

Commissioner Kathleen Casey:
Thank you very much, Robert. I want to go back to an earlier reference by Mr. Rosenblatt and Sachs, with respect to, again, the focus on whether or not our market structure is actually supporting a capital formation function.

And one of the questions I have is that some of the commenter’s — and think the [unintelligible] — is that when you look at some of the performance metrics, which have been noted today, in terms of spread, volumes, liquidity, costs, and the benefits that I think that we’ve been able to identify in terms of the tightening of spreads, increased volume, increased liquidity, and the reduction of costs, how concerned should the Commission be, or focused should the commission be on the distinctions that can be drawn between smaller cap and larger cap companies, and whether or not market structure is sufficiently supporting smaller cap, less-frequently-traded companies? And that what that may mean for capital formation?

Joseph Ratterman:
I’ll take a quick stab at that, that just to go on record and say that I don’t believe today’s market structure works equally well for stocks — for the large, highly liquid stocks as it does for lower-volume stocks. I don’t know that we have an answer there. I think that somewhere in there, some level of competition incentives will provide part of the answer, but I think it’s clear today that what does work well for the top 1,500 or 2,000 stocks does not work well, especially in times of stress, for the bottom 1,500 or 2,000 stocks.

Richard Rosenblatt:
I think I don’t have a direct answer; I do have an indirect one. I think we spoke earlier about limit orders. I’m a believer in market orders because a market order is an expression by an investor that they have confidence in their price discovery mechanism of the marketplace that they sent that order to. It’s not that they’re willing to buy stock at a million dollars or sell it at zero; it’s that they believe the market will work efficiently and that their order will be priced at a reasonable price.

When that falls apart, it’s not the fault of the market order, but the marketplace that received it. I think that there’s no question that our markets for small caps are nowhere nearly as efficient, nor would you expect them to be, as for almost liquid names.

However, when we’re discussing capital formation, the job of secondary markets is to instill confidence, to let investors know that when they do decide to make in investment change, it will be accommodated in a fair and efficient way. I think in dealing with our general market structure, to plug loopholes, to continue to improve so that investor confidence is maximized, we will aid in capital formation and contribute to better markets in small cap names.

Joseph Ratterman:
If I could [unintelligible] — one quick comment in response. I feel very strongly against the use of market orders. Not to pick a fight on the panel, but it’s the only market system where users can send in a transaction request to execute at any price. When I go to buy a car or I go to the grocery store, I have a menu, I have a list of prices that I’m expecting to get and I will get that based on the transaction guarantee. In the U.S. equity markets, I feel strongly customers should price their orders and give us their intent. It’s a dangerous order type that doesn’t apply in hardly any other market that I’m aware of.

Ian Domowitz:
If I could raise the capital raising flag to separate the combatants for just a second and go back to that question. It’s an extremely intuitive one, actually, and very important, although I do not believe it rests with small cap securities, although obviously, that is, in some sense, where every company starts. The analysis of transactions costs has revealed astounding anomalies with respect to large cap versus small cap, vis--vis one’s expectations. I won’t dwell there.

Your main point was capital-raising, all right? And in the academic literature, and indeed, in the industry as a whole, this has everything to do with the so-called “cost of capital”. And it’s certainly true that we tend to write much more on the secondary markets as opposed to the primary, in the sense or the connection between the two, given the difficulty of doing so. I can, at request, provide some references.

But I can tell you what the general tenor of that discussion in that — at least that research is. And that is that to the extent that market structure reduces transactions costs, regardless of class or security, it is demonstrable — or at least, has been demonstrated empirically — that the cost of capital is reduced. And one can actually work out, all right, from the numbers the notion of any elasticity if you like; you know, there is the degree to which transactions costs must fall in order to get the cost of capital to fall a certain amount or vice versa, obviously.

So, that connection actually has been made. It ties directly to market structure only insofar as market structure affects the cost of transacting, all right, and therefore, also ties to your comment about small capitalization stocks.

Charles Jones:
Joe is right that the small cap stocks — it’s not working. The new market structure doesn’t treat these the same way as it treats large liquid stocks.

But the question really is what would the alternative be? What would — how does it compare to what we had before for small cap stocks? And I think the evidence on that — I’ve written a paper with Terry Hendershott, who’s on a later panel, and maybe we can come back to this — but essentially, there seems to be no harm in any of the rule’s changes that have come along. So, it looks like the small cap stocks haven’t benefitted from the new market structure, but neither have they been hurt. So, in a sense that they’re — they — we haven’t helped their case, but we’ve really made things better for the large cap stocks. But there doesn’t seem to be any evidence of damage to the liquidity based on the current market structure.

Stephen Sachs:
Ken, if I may — I was just going to say if I may continue on that point anecdotally from a trader’s perspective. Interestingly enough, trading small cap stocks today is really not vastly from the way I traded them 15 years ago. All the things that we’ve been discussing here today, particularly as it relates to the incentives in the decimal world, in the electronic world of the quote and flickering of quotes, and, you know, highly liquid, high-volume stocks, the small cap, and to a great extent, the mid cap arena, has been vastly left behind there.

So again, from an investor or trader’s perspective and actually having to execute large orders in these small cap stocks, I would completely agree that they have not been advantaged nor disadvantaged. It really is, you know, akin to trading in these stocks, you know, 15 years ago. So, they don’t have the noise level, if you will, and ultimately — you know, right, the press has talked about this a lot — from a fundamental investment perspective, there’s no good reason why Citigroup trades a billion shares a day. The fundamentals of Citigroup don’t change that quickly. So, if you compare that or contrast that with a small cap security, the trades may be 100,000 shares a day, the argument — I think the long term-capital formation perspective is that is a far more efficient market. You are actually getting the fundamental investor’s true viewpoint of price discovery.

Christopher Nagy:
I would like to add just one quick point to that. I do believe that small cap capital formation has been harmed, because quite frankly, we have not see a lot of new small cap companies come to market in quite some time, which retail investors favor quite a bit.

I will say, and this is probably best suited for your next panel, that we’ve noticed that a lot of the newer trading algorithms and newer trading styles tend to circle around the more active securities in the marketplace, like the Citigroups, where they’re completely devoid of any sort of activity in these smaller cap stocks, and again, we think that can be cured in large part by incentivizing capital commitment in those stocks. And if you, you know, incentivize electronic market-makers or specialists to commit capital in those stocks, I think you’d have much more robust structure, and i.e., potentially, more capital formation in the U.S. as a result.

Timothy Sargent:
Excuse me. Our experience also, you know, belies that change, but it’s because many times they’ve experimented with new technologies that are appropriate in that, you know, for those larger, more liquid orders, and when they go to use those technologies in that space, it’s a very disappointing result. They go back in with participation strategies that are essentially well understood in those larger areas. They tend to be more worried again about that information signaling as they’re consistently, you know, showing up in these names and drawing attention to themselves, and probably prefer that more upstairs-negotiated kind of liquidity activity that is more consistent with the way things were handled before.

Gus Sauter:
I think the lack of new issues is probably more a function of what’s been going on in the economy over the last couple of years than anything else. If you have a true long-term investor who wants to buy and another long-term investor who wants to sell, the ideal market structure is one that gets them together and enables them to do it in a cost-effective fashion, or with no cost, ideally. And I think that’s what the electronic markets and the current market structure in total has accomplished for securities whether there are enough people that want to buy and sell at the same price, so they basically remove the middleman.

Unfortunately, in the smaller cap area, there just aren’t as many opportunities to cross natural investors, so you do need intermediaries, and so there is a cost to intermediation, but yet an advantage to it as well, providing liquidity. So, I think that’s why we’ve seen the dramatic reduction in transaction costs in the large cap area, because the market structure has enabled that, but there isn’t the natural liquidity in the small cap segment.

Director Robert Cook:
We’re nearing the end of our time. I wanted to leave time for any follow-up questions that the commissioner —

Commissioner Luis Aguilar:
One question, and I recognize we’re running out of time, but I would like to get your views on the trade-out concept. I know from reading your materials there are some disagreements as to the pros and cons, and before we run out of time I would like to get your views on those.

Gus Sauter:
Well, I guess I stated a couple of times that we do favor the trade-at rule. We think that if you don’t have a trade-at rule, it disincents people to put the orders on the book. If they know that somebody can participate ahead of them despite not having displayed liquidity, then why display your own liquidity? So, that’s the primary reason we do favor the trade-at rule. We think that a first-come, first-serve type of arrangement should be honored.

Christopher Nagy:
We do not favor the trade-at rule, and it’s primarily because of the cost considerations in the marketplace. So, if you take a look at the inputs of execution costs in the marketplace, we see that those costs would potentially go much higher with the trade-at rule than they would today. So, unless there were some measures to curb costs in that consideration then we think it would harm individual investors.

Male Speaker:
Can I ask why that would be? I mean, basically, you would just have to route to another exchange and still obtain the same costs.

Christopher Nagy:
Right. You’d have to route to exchange, necessarily, where you’d incur a 30-cent-per-hundred charge, where, in today’s markets, you’d be able to use an agency or an internalizer to lower that execution cost, and potentially even run into a negative cost situation. And so there are cost considerations in the market that need to be looked at, and I would surmise that’s exactly why we have Rule 610 in the equity marketplace and why is 610 is now being proposed in the options marketplace: because of those cost considerations.

Richard Rosenblatt:
I think it’s important that we be very concerned about orders that are not exposed to markets, and there are good economic reasons why that should occur; Chris is pointing out cost benefits.

However, we’re all in this business to make money, and your job is to make sure the public is protected while we’re making money. I think the trade-at rule, or something similar, speaks to that obligation. If in fact an order is to be withheld from the marketplace where it will achieve an equal or better price, there has to be a good reason for that. If price improvement is being offered, it should be significant price improvement. If cost advantages are being offered, well, they should be demonstrable.

I think, as I said before, it’s very important to incent transparency in our markets. I think it’s essential if Reg NMS is going to be as successful as we all hoped it would be when it was presented. And the trade-at rule is something that does incent transparency, because now I know that if I’m bidding or offering in a public market, that no one will trade at my price, or a worse price, without sending — well, it wouldn’t trade at a worse price anyway — but no one would trade at my price without sending me that order first.

So, it’s certainly something that we do need to consider because I think one of the biggest problems we face is incenting limit orders on our books.

Christopher Nagy:
I would agree with incenting limit orders, but there’s one other fear of ours that I would caution the Commission on, and that’s as it relates to information leakage.

We’ve talked about the small size in the marketplace today. Our actual average execution order size is 1,500 shares, believe that or not, and if you look in the markets today, with the small order size, if we were forced with the trade-at rule — imagine this — that we go out and we get the hundred shares that were offered at that trade-at price, what would the remaining executions fall in at? Where would those prices be? What would happen to high frequency traders’ ability that’s co-located to immediately send cancellations in on their orders? We see signs of that happening today.

So, our concern isn’t necessarily at the price but it’s that size at the inside and what happens to the downstream effects of that order with the leakage that’s going to happen on the information when the order goes in. So, that’s another consideration, as well, too.

Joseph Ratterman:
We feel like the trade-at rule is definitely a solution aimed at the right problem. We don’t know if it’s exactly the right solution; it feels a little heavy handed to me.

There is an underlying issue that we’ve looked at based on a lot of data that shows that most of the executions in dark markets are in lower priced stocks. While it’s not a popular proposal, another solution is to look at allowing the displayed markets to better represent the fair value of the underlying security.

So, there’s a lot that goes along with that particular solution in terms of more market data and potentially what appears to be insignificant price points for the retail investors, specifically. But it is another way of looking at the problem, which is that the displayed markets may not have the ability to truly represent the underlying value of the securities, allowing for dark markets to ride on top of that price discovery at maybe arbitrarily wide-price increments.

Director Robert Cook:
Thanks Joe. Commissioner Walter?

Commissioner Elisse Walter:
I wondered if you could quickly, in the time we have, give me a reaction to internalization and whether you think that’s a privilege that should come with certain obligations and how that fits into the other topics you’ve been discussing?

Timothy Sargent:
I could just make a comment, because many of our clients have now reached out on their own to be able to gather those information sets because they’re not naturally available. And almost to a client, anecdotally, whenever they do measure, they tend to have poor executions and use those destinations less frequently.

Gus Sauter:
I can tell you that we get very frustrated with internalization. There’s nothing more frustrating than being willing to execute a transaction or trade and have yours not go off and see the trade go off elsewhere because of internalization. Internalization uses market information provided by other venues and without rewarding those venues. So, we think it locks up liquidity which would be better off participating in a broader marketplace.

Stephen Sachs:
Yeah, I would comment that our experience has been very similar, and that generally speaking, our executions on internalized orders are among our, you know, lowest quality, if you will, or providing, generally the highest signal, you know, from what we have found.

I think it ultimately, you know, dovetails right into the trade-at, you know, argument and that, you know, again, I generally support the trade-at rule. I think that the advantages outweigh the disadvantages because all of this is intermingled with what we’ve been discussing here today from an incentive, you know, perspective. But generally speaking, I think the trade-at and the internalization practices I think would need to be tied, you know, very closely together.

Director Robert Cook:
Thank you. With that, I think in the interest of time, we should bring this first panel to a close. I’d like to thank all the panelists for sharing their time and expertise and insights with us today. It’s been a very useful dialogue, and I’m sure we’ll be continuing with the discussion on many of these issues as we move forward.

I think we have a five-minute break before the next panel convenes. Five to ten minute break? Okay, thank you.

[end of transcript]

 

http://www.sec.gov/news/otherwebcasts/2010/060210marketstructure-trans1.htm


Modified: 06/18/2010