1 MEETING 2 of the 3 MARKET STRUCTURE HEARINGS 4 New York University 5 Tisch Hall 6 New York, N.Y. 7 November 12, 2002 8 9:00 a.m. 9 ANNETTE L. NAZARETH, Chairperson, Presiding 10 DIVERSIFIED REPORTING COMPANY, INC., 11 1025 Vermont Avenue, 12 Washington, D.C. 20005 13 (202) 296-9626 14 15 16 17 18 19 20 21 22 23 24 Diversified Reporting Services, Inc. 25 (202) 467-9200 1 P-R-O-C-E-E-D-I-N-G-S 2 COMMISSIONER GLASSMAN: Good morning. I'm 3 Commissioner Cynthia Glassman, and on behalf of my 4 colleagues, Chairman Pitt, Commissioners Goldschmid, Atkins 5 and Campos, who will be here in a second, I'd like to welcome 6 all of you today to our market structure hearing, and thank 7 you all for your participation. 8 Our first interactive session in Washington, on 9 October 29th, was very informative, and I know I speak for 10 all of my colleagues when I say that we have been looking 11 forward to the second session today. 12 As I said at the last session, the purpose of 13 today's session is to educate the Commission on issues 14 relating to the structure of the U.S. equities markets, with 15 emphasis on how the public interest and protection of 16 investors can best be served. 17 A key subtheme is whether our well-intended rules 18 to protect the public have unintended consequences. We want 19 the benefits of your views on the topics we've chosen for 20 discussion today. Even more useful to us, will be the 21 opportunity to listen to you, the panelists, discuss and 22 debate this array of issues. 23 In my ten months as Commissioner, I've had meetings 24 with representatives of many different markets and market 25 participants, and each meeting has been useful. But what I 1 really wanted to do was to get all of the differing 2 viewpoints on a particular topic represented in the same 3 room, and let the people advocating one position respond to 4 those advocating a contrary position. No holds barred. 5 We had some lively discussion last time, and this 6 is what we hope will happen today. 7 As far as our expectations, it's obvious that two 8 days of hearings, no matter how productive they are, are not 9 going to answer all the questions. But today's discussion 10 will enhance our understanding of current market structure 11 policy concerns, as well as possible tradeoffs that we may 12 have to make as we attempt to resolve the important issues 13 before us. 14 Before we get started, I'd like to again thank the 15 key members of our market structure working group. Annette 16 Nazareth, Bob Colby and Alden Atkins and their staff in 17 market reg. Larry Harris, our chief economist and his 18 colleagues; Giovano Presioso, and Meridith Mitchell, who are 19 not here today, but from our general counsel's office. Steve 20 Jung on the Chairman's staff and last but not least, Mary 21 Head and Brian Stern on my staff. 22 Now, let's get started. I'll turn this over to 23 Annette. Thanks. 24 MS. NAZARETH: Thank you, Commissioner Glassman. 25 I, too, would like to welcome the Commission, our 1 distinguished panelists and members of the public to this 2 second market structure hearing. Today the Commission 3 continues its more than thirty-year tradition of critically 4 examining issues relating to the structure of our equity 5 markets. Prior Commissions have faced many complex and novel 6 issues in this particularly challenging area of securities 7 regulation, both before and after the Securities Act 8 Amendments of 1975 that gave the Commission a mandate to 9 build a better national market system. 10 To its credit, the Commission took actions that 11 have resulted in a vibrant, healthy and efficient 12 marketplace, a marketplace indeed that is the envy of the 13 world. As technology continues to transform our lives and 14 the way we do business, once again we stand at a crossroads 15 with respect to various market structure issues, and the 16 solutions we arrive at will have far-reaching consequences, 17 especially in this highly competitive marketplace. 18 We hope that these hearings will achieve several 19 important objectives. First, as Commissioner Glassman 20 mentioned, we hope that these hearings will provide a 21 comprehensive picture of the structure of the U.S. equities 22 markets. Our marketplace consists of a sweeping variety of 23 market centers and trading methods that cater to different 24 types of securities and different categories of investors. 25 Some markets are centralized, others are decentralized. Some 1 markets are driven by the interaction of customer orders, 2 others are propelled by the willingness of designated market 3 intermediaries to committed capital. Some markets are 4 grounded in the physical trading floor where bids and 5 offerings meet face-to-face. Others float in the anonymous 6 realm of cyberspace. 7 Another goal of these hearings is to discuss the 8 complex tradeoffs that we have to make in regulating market 9 structure. How should we reconcile the investor's need to 10 obtain current information about market activity with each 11 market center's desire to exploit the commercial value of the 12 data that it generates? How should we reconcile the 13 investor's desire to have access to the best prices with the 14 right of individual market participants to employ the 15 business model of their choice? 16 When should a market be required to act in the 17 public interest, rather than in the interest of its members 18 or shareholders? 19 These questions do not have simple answers, and 20 there are many difficult decisions to be made. We are 21 nevertheless committed to resolving these competing interests 22 in the fairest and most balanced manner possible. 23 A third goal of these hearings is to use what we 24 learn to maintain consistent policy positions on the key 25 issues facing the Commission. In recent years, the 1 Commission has undertaken a number of market structure 2 initiatives. Just to name a few, the Commission adopted 3 order handling rules for market makers and ECNs in 1996, 4 Regulation ATS governing alternative trading systems in 1998, 5 and order execution and order routing quality disclosure 6 obligations for market centers and retail brokers in the year 7 2000. 8 While some may differ, most would say that these 9 rules have significantly improved the transparency and 10 accessibility of our markets. 11 The Commission has also solicited comment on a 12 number of specific areas of market regulation. Among others, 13 the Commission has issued concept releases on the regulation 14 of market data fees and revenues, the fragmentation of the 15 U.S. securities markets, and the regulation of exchanges. 16 The Commission also convened an advisory committee in 2000, 17 chaired by Dean Joel Seligman, to consider issues related to 18 the collection and dissemination of market information. 19 These efforts to develop sound policies for market 20 regulation must continue if we are to make our markets 21 fairer, more efficient, and more transparent. To this end 22 the Commission seeks to ensure that the decisions it makes in 23 the coming months and years will benefit from the wise 24 counsel of the industry, the academic community, and the 25 public. 1 Now, before I introduce our participants today, 2 I'll briefly mention just a few housekeeping matters. This 3 morning we will have two sessions. In the opening session, 4 we'll touch on the overarching principles of market structure 5 and discuss whether there is a consensus on those guiding 6 principles. This session will last only an hour, and will be 7 followed by a fifteen-minute break from approximately 10:30 8 to 10:45. We may run a little bit late. 9 We will then follow with our first major session on 10 market data to be moderated by Bob Colby of the Division of 11 Market Regulation. At 12:15 we'll break for lunch for one 12 hour and then reconvene at 1:15 for our afternoon program. 13 As to the general groundrules, in order to keep the 14 dialogue moving and to afford everyone an opportunity to 15 speak, and I am very optimistic, I'm told we have a group of 16 good chatters here, I'd like to ask each of you to attempt to 17 limit your responses to two or three minutes, and given the 18 limitations on time, we also ask that you not pipe in just to 19 repeat a point that's already been made. 20 The next groundrules are just as tough. Try to 21 leave your ego at the door, your own personal business 22 perspectives and biases, and keep in mind that the 23 Commission's ultimate goal is the protection of investors; 24 that this is a policy discussion after all. And I'd also 25 like to note that just outside this room, we have a black 1 box, which you can put note cards with questions that you may 2 have for the panel that, time permitting, we will attempt to 3 address later in this program. 4 So with that underway, I'd like to introduce each 5 of the participants briefly and then circle back and give 6 everyone the opportunity, if you wish, to make a brief 7 statement, although it's not required, but you're certainly 8 welcome to. 9 Before I do that, I would like to introduce our 10 Commission which is present today. We have Chairman Pitt, 11 Commissioners Goldschmid, Atkins and Campos and of course 12 Commissioner Glassman. I'd especially like to thank Chairman 13 Pitt for calling for these hearings and Commissioner Glassman 14 for her very substantial efforts in putting this program 15 together. 16 We also have at the table Bob Colby, who is the 17 Deputy Director of Market Regulation, and Larry Harris, who 18 is our chief economist. 19 Going around the table, in order, we have Robert 20 Murphy, who is here from LaBranche. We have Commissioner 21 Campos. We have Albert Kyle, who is a professor from Duke 22 University; Minder Cheng, who is with Barclay Global 23 Investors, Ian Domowitz, with Investment Technology Group; 24 Janet Angstadt with Archipelago; Matt DeSalvo with Morgan 25 Stanley; Scott DeSano from Fidelity Management and Research, 1 I think the staff is pulling a stunt on me, DeSalvo and 2 DeSano next to each other. Larry Harris, Chairman Pitt, Bob 3 Colby, Commissioner Glassman, we have Tom Peterffy from 4 Interactive Brokers Group, Tom Gardner, who you may not 5 recognize without his hat, here from Motley Fool, Paul 6 O'Kelly from the Chicago Stock Exchange, Robert Steel from 7 Goldman Sachs, Gunner Burckhart with Deutsche Asset 8 Management; Commissioner Goldschmid, Commissioner Atkins and 9 Rubin Lee from Oxford Financial Group. 10 I'd like again to give each of you the opportunity 11 to make a brief statement. We could start, Bob, with you, if 12 you'd like to. 13 MR. MURPHY: Be happy to. Good morning, I am Bob 14 Murphy, New York Stock Exchange specialist with LaBranche & 15 Company as well as New York Stock Exchange Director and Vice 16 Chairman of the board. I wish to thank the Commission for 17 affording us this opportunity to exchange views on issues 18 relating to the structure of our security markets. 19 Our markets can only benefit from these hearings. 20 Specialists on the New York Stock Exchange are accountable to 21 the investing public, for the quality of NYSE markets and 22 their specialty stocks. This means that the specialist is 23 responsible for fostering and maintaining liquidity and 24 continuing two-sided auction markets on the NYSE floor. The 25 specialist plays a pivotal role in insuring that investor 1 orders receive the best possible execution. To accomplish 2 this, specialists act both as agent and principal in their 3 specialty stocks, to help insure that markets for those 4 stocks are fair and orderly, and that they operate 5 efficiently in the public interest. 6 The NYSE specialist is the focal point of the 7 auction price discovery process. The specialist's primary 8 job is to bring buyers and sellers together in an efficient 9 and orderly manner to provide transparency and liquidity and 10 to maximize the opportunity for public buy and sell orders to 11 interact without getting between them. 12 The NYSE specialist has an affirmative obligation 13 to step in and step up when no one else wants to. The 14 specialist buys or sells as principal when necessary to 15 minimize short-term imbalances between supply and demand. 16 In this way, the specialist provides depth and 17 price continuity, minimizes volume at this time in the 18 auction market. In addition, the specialist has a negative 19 obligation. The specialist may not trade for a proprietary 20 account unless reasonably necessary to maintain a fair and 21 orderly market. 22 The NYSE specialist places customers on both sides 23 of the market first. In my view, all of us in the industry; 24 brokers, dealers, market makers and specialists, should 25 adhere to that fundamental principle. 1 The industry has changed quite a bit since I first 2 came to Wall Street and not all for the better. I am 3 concerned about a number of anticustomer practices that I see 4 in the industry today. Dealers in other markets cherry pick 5 the easiest orders of the least informed customers and 6 execute those orders at prices discovered at the NYSE while 7 NYSE specialists stand ready to accept all orders of all 8 customers, including the offsetting orders of the cherry 9 pickers. 10 Agents for customers receive payment for their 11 customer orders. Agents for customers trade against orders 12 for their customers, without giving those orders a chance to 13 interact with other orders of public customers. Dealers in 14 other markets disseminate non-competitive quotes, thereby 15 shunning the responsibility to make markets for all, while 16 still guaranteeing significant order flow. An exchange buys 17 reports of off-exchange trades from OTC dealers and ECNs and 18 fraudulently prints those off exchange trades as though they 19 have actually taken place on that exchange. An ECN hides 20 quotes and its participants ignore better priced orders on 21 other markets. 22 Regaining investor confidence is crucial to the 23 continued preeminence of our secondary markets, yet we have 24 allowed to creep into our markets practices in which dealers 25 and even agents put their own interests ahead of the 1 interests of their customers. 2 The NYSE trading model precludes these 3 anti-investor practices by the very way that it has evolved. 4 We integrate the trading of institutional and retail orders 5 into a single auction and we apply automation to the extent 6 consistent with that integration. This provides investors 7 with optimal price discovery and maximum flexibility. Other 8 markets take different approaches, but whatever model a 9 market may choose, the Commission should not permit it to 10 embody those anti-investor practices -- and that requires the 11 Commission to attack the root causes of those practices. 12 Finally we believe Congress got it right in 1975 13 when it determined that transparency was the heart of 14 effective inter-market competition and when it bounded 15 intermarket competition by the principle that dealers should 16 only intervene as a last resort, thus, we support all markets 17 disseminating as much as is possible, trading interests 18 disclosed at the point of sale. We support maximizing the 19 opportunities for customer orders to interact directly within 20 a market so that the particular participants to a trade 21 receive best execution and all market participants receive 22 the benefits of optimal price discovery. 23 Let's place the customer first. I thank you for 24 this opportunity to present my views. 25 MS. NAZARETH: Thank you, Bob. Pete? 1 PROFESSOR KYLE: Thank you. I just want to make a 2 couple of points. Three points. First, computers have 3 become very commonplace in trading and we need to think about 4 what the role of computers is. On the one hand, computers 5 allow us to implement very complicated trading strategies and 6 keep track of very large amounts of information. On the 7 other hand, computers are very honest. If they do something 8 wrong you can generally catch them and they generally do 9 whatever you tell them to do and we shouldn't lose sight of 10 the fact that computers will allow us to have a better audit 11 trail and more continuity in the markets. 12 Second of all, I want to make a few points of 13 organized exchanges. Organized exchanges have sophisticated 14 members and sophisticated customers. They also have 15 unsophisticated customers who are usually retail type people. 16 The value of an organized exchange's franchise largely lies 17 with its reputation among the least sophisticated customers. 18 And the ability of an exchange to build this reputation 19 largely consists of how transparent the exchange is, so the 20 idea of building transparency is very important for organized 21 exchanges to maintain their reputation, which is their 22 capital, with the least sophisticated customers. 23 The third point I want to make, we have with the 24 national market system a ticker, and that ticker could easily 25 be augmented to have added to it various pieces of 1 information that would make the market transparent and also 2 make it easier to audit what's going on in the market. Among 3 other things, you could have a unique identifier attached to 4 a ticker that would allow it to be traced back to a 5 customer's trade so a customer can easily identify his trade 6 on the ticker. Whether you're a large customer or a small 7 customer, that type of information would be very valuable and 8 would not be all that costly to implement. 9 Thanks. 10 MS. NAZARETH: Thank you. Minder? 11 MR. CHENG: Good morning. I also would like to 12 thank the Commission for giving us the opportunity to comment 13 on these important issues that will be important not only to 14 us, but also to the market and to the investors. 15 Let me just start by saying I think that trading, 16 especially on the buy side for the large institutional asset 17 managers has evolved from pure execution to an element of -- 18 a cost element of the investment return to now account for a 19 large portion of the additional return we can provide to our 20 investors. 21 Now, with the technology and with regulatory 22 environment changes, it has opened the way for buy side 23 traders to directly access market liquidity, but what it also 24 has introduced is it created an environment where it becomes 25 a challenge for the buy side institution to find a way to 1 additional return, minimize transaction cost, but then 2 without really going through some of the hurdles that 3 certainly did not exist before. 4 And I would just highlight one major challenge, I 5 think, which I think will be discussed later, for a buy side 6 trading operation, generally speaking, the most difficulty 7 challenge is to move large blocks of orders through the 8 market over multiple days. So we're going to get into a 9 discussion of fairness and transparency in the market. From 10 our point of view, meaning for at least for a buy side shop, 11 transparency generally is a good thing, but may not be too 12 good a thing, given what we're trying to do. 13 In other words, this may certainly introduce some 14 bit of a controversy, so I was hoping to see an environment 15 where there's enough transparency, enough fairness, but 16 without compromising buy side trading's capability to get our 17 trades done without paying excessive market impact cost. 18 Thank you. 19 MR. DOMOWITZ: Thank you for inviting me here 20 today. I join with everyone else in the most interesting 21 times in market structure and possibly all history. 22 I'm the managing director of products at an 23 institutional agency brokerage, and as overseer of three 24 alternative trading systems, I obviously believe that ATS 25 systems provide value and contribute to best execution for 1 our clients. Our own system is non-price discovery in 2 nature, attests to a deep interest in a market structure that 3 promotes meaningful price discovery in the national market. 4 Now, as such, I therefore promote consolidation, 5 but I don't believe that it ought to be mandated or 6 necessarily all encompassing, and in thinking about market 7 driven solutions, I think that we all must also recognize 8 that exchanges now compete with brokers along with a variety 9 of dimensions. This complication has implications which I 10 believe should be acknowledged and I'm sure will come up here 11 today. 12 Now, as I say, ITG is an agency brokerage and we do 13 not commit capital. Nevertheless, such commitment is very 14 important and should be fostered in a competitive 15 marketplace. In today's environment, however, I don't 16 believe that capital provision need be tied to any particular 17 form of market microstructure. In other words, if immediacy 18 is indeed demanded, it will be supplied at some price under a 19 variety of conditions. 20 Finally, I view best execution which I believe is a 21 topic that will come up today, not just as a set of reporting 22 requirements, but as a process. And these are broker 23 responsibilities, by and large, and they include measurement 24 analysis and control of transaction costs on the part of 25 clients. 1 I believe in development of market structure that 2 contributes to and perhaps more importantly, may depend on 3 best execution obligations, and I hope that some of our 4 discussion focuses on that today. 5 Thank you again. 6 MS. NAZARETH: Janet. 7 MS. ANGSTADT: Good morning, I'm Janet Angstadt 8 here for Jerry Putnam. I am with Archipelago Holdings and I 9 think we're uniquely positioned to talk about market 10 structure from the ECN and the market exchange perspective. 11 We've been an ECN for five years now and an exchange since 12 March, and we've really benefited from, a lot of innovative 13 mechanisms that the SEC put in place; the order handling 14 rules that got us to the quote, the Reg ATS that gave us 15 some certainty as to our status as an ATS and also invited us 16 to become an exchange. 17 I think from our perspective, we found that in fact 18 the state of the union is very good. We've seen a lot of 19 good initiatives from the Commission, again the order 20 handling rules, Reg ATS, decimals 1-5. There are just a 21 couple of areas that we should see some attention should be 22 given, and in fact those things are happening now. I think 23 with respect to access among markets and making sure that 24 access between these competing markets is fair. If that's in 25 the ATS arena, let's make sure that access fees are in fact 1 fair between these competing markets, and with respect to 2 linkage, between the reform that the Commission has urged 3 with the ITS linkage we think that's a good area for these 4 market areas to really push to bring some innovation and some 5 technology, so we would ask the Commission to watch closely 6 what the ITS operating committee is doing there and make sure 7 there's enough momentum for that group to continue, but we're 8 pleased to be part of the debate this morning. 9 MS. NAZARETH: Matt DeSalvo? 10 MR. DeSALVO: Matt DeSalvo, Morgan Stanley. I come 11 here as a broker-dealer that represents order flow that 12 originates from both the retail and institutional side on 13 both an agency and principal basis; that transact in the 14 public marketplace on both the auction, electronic and ATS 15 level in both cash and derivatives. 16 One of the biggest structural issues facing the 17 market today is relevant inaccessibility of prices in the 18 system often driving the NBBO. The concept of fairness is 19 paramount in our mind also. We need to end up with either an 20 access standard for all markets that require quick execution 21 against split interests. We offer a flexible approach to 22 best execution, recognizing the fact that there are actual 23 and opportunity costs associated with trying to reach prices 24 in non-automated and unlinked markets. Whatever the SEC does 25 in terms of market structure at the end of the day, its 1 paramount objective should be levelling the playing field. 2 That means eliminating or rationalizing access fees. In 3 other words, this issue right now where market makers and 4 ECNs have one ability to charge and one not to charge needs 5 to be addressed. That means answering the question as to why 6 broker-dealers have to pay exorbitant rates to the monopoly 7 for market data that it needs to satisfy, -- to meet best 8 execution requirements. That means access standards and best 9 execution obligations that could be fairly and realistically 10 administered. 11 MS. NAZARETH: Scott. 12 MR. DeSANO: Good morning, thanks for the 13 opportunity to be here today. I'm Scott DeSano from Fidelity 14 Investments. Fidelity along with the fund industry generally 15 serves individual investors. Mutual funds as a whole have 16 over 6 trillion in assets under management, which it 17 reinvests for over 90 million individuals. As of October, 18 Fidelity had over 400 billion in equity assets in over 200 19 equity funds domestically. 20 We approach market structure issues with only one 21 purpose in mind and that's to better the returns for our 22 funds and in turn for the individual investors who invest in 23 our funds. For this reason, distinctions between retail 24 investors and institutional investors may not always be the 25 most useful. Through full and fair competition, investors 1 are more likely to achieve best execution for their trades, 2 but best execution is in the eye of the beholder. 3 The goal of best execution is not limited to 4 obtaining the best price. As important, if not more 5 important, is the speed and certainty of execution. This is 6 especially the case for trading the large blocks of stock 7 given the volatility of the markets and the demands of 8 liquidity involved in trading large blocks of stock. 9 Liquidity is very key. 10 Market centers should disclose their entire limit 11 order book. The securities markets have begun to make 12 improvements in this area with Super Montage and the New 13 York's Open Book, but these steps fall short of full and 14 timely transparency. Disclosing the size and price levels of 15 orders on the limit order book will lead to greater liquidity 16 and depth of trading. 17 Mutual funds and other institutional investors will 18 add to liquidity by placing their orders on the limit order 19 books if strict time and price priority are given to those 20 orders over orders entered later on the book or made in the 21 trading platform. The lack of strict time and price 22 priorities effectively deters institutional investors 23 including Fidelity from placing limit orders on the 24 specialist books. This contributes to fragmentation and a 25 lack of transparency in the market and detracts from the 1 depths and liquidity in the trading of stocks. The lack of 2 strict time and price of trading of orders has undermined the 3 appeal of the New York institutional express system for the 4 trading of large blocks of securities. The practice of being 5 pennied by traders on the Floor adds to the disincentives 6 against placing limit orders on the specialist books. 7 Decimalization was intended to foster better 8 execution for customer trades by narrowing spreads and adding 9 liquidity to the markets. The unintended consequence of 10 decimalization has in fact made best execution problematic 11 for investors. Traders and specialists are now able to jump 12 ahead of earlier entered customer orders with much less 13 downside risk. They can offer minimal price improvement of 14 one cent to the ordering sides of the trade with the 15 knowledge that if the market turns against them they are 16 likely to lay up their position with little loss while able 17 to capture far greater upside gain. This forced us as an 18 institution to trade predominantly upstairs or through ECNs. 19 It is open to serious question whether on balance investors 20 obtain better execution within an environment that provides 21 floor traders with heightened trading advantages over 22 customer limit orders. 23 Thank you. 24 MS. NAZARETH: Tom Peterffy. 25 MR. PETERFFY: I'm Tom Peterffy, Chairman of 1 Interact Group. We are an electronic broker-dealer who 2 provide liquidity and market makers and execute for our 3 direct access profession accounts. We do this fully 4 electronically in Asia and Europe and to the extent possible 5 in the United States. 6 We appreciate this opportunity to tell you our 7 views. We think that broker-dealers should be able to 8 interact with electronically accessible limit orders. We 9 should provide strong economic incentives to traditional 10 exchanges throughout the world. Approximately 100 of our 11 employees work as market makers on the floors of the four 12 traditional option exchanges and two oversee our market 13 making activities on the fully electronic ISE option 14 exchange. These two people transact more volume every day 15 than the other 100 combined. We need electronically 16 accessible limit order books with automated matching to 17 efficiently fulfill our duty of best execution and provide 18 access to the markets. Thank you. 19 MR. GARDNER: I'm Tom Gardner, co-founder of Motley 20 Fool. I'd like to thank the Commission for including me and 21 the Fool today in the proceedings. 22 We are an organization that reaches more than 20 23 million investors each month through our on line service, 24 newspaper column and our radio program on NPR. I am not 25 someone who works at a firm that handles orders and we are 1 not an organization that speaks to active traders that are 2 using many of the trading systems that we'll speak of today, 3 so I am largely irrelevant to today's proceedings and in the 4 interests of full disclosure I want to get that out there. 5 I can speak about general philosophy, and that is 6 that we believe that healthy markets, a rise in liquidity, in 7 trading activity rely on equal access to material information 8 and presently available prices in the national market system 9 for all investors, regardless of the size of their account 10 and regardless of the size of their order. We do not believe 11 that distinctions between retail institutional investors are 12 useful or should be relevant to the work, those regulating 13 the marketplace. However, we should not lose sight of the 14 fact that our markets in large part because of the SEC over 15 the last decade are increasingly transparent, efficient, more 16 competitive and more accessible to all investors, but 17 progress requires vigilance. At this point, henceforth, 18 today I am the quintessential proxy for Dr. Kyle's least 19 sophisticated customer. 20 MS. NAZARETH: And proud of it. Paul. 21 MR. O'KELLY: Thank you, Annette. First of all, 22 I'd like to commend the SEC for undertaking this review of 23 market structure and thank them for allowing the Chicago 24 Stock Exchange to participate today. 25 More than 25 years ago, Congress directed the SEC 1 to facilitate the creation of a national market system. When 2 it met the objectives that Congress identified as important 3 for protection of public investors, one built on a set of 4 principles that Congress also identified in the 1975 Act 5 amendments. To those who listened to the market structure 6 hearings in Washington a couple of weeks back, you heard 7 Annette ask the various participants whether those principles 8 continue to be relevant today when determining how to 9 structure a market. It was one of the few areas where there 10 was a consensus among the participants. 11 I mention this fact, because when the Commission 12 market participants back in 1975 built the national market 13 structure system, it contained two principal attributes. One 14 was the consolidated mandatory display of quote and trade 15 information to promote market transparency, and the other was 16 intermarket access to prices which promote the best execution 17 of orders. I think that if the Commission again examines 18 market structure in light of these principles, it will reach 19 the same conclusion, and the market will look as it does 20 today at the current national market structure. The 21 challenge for the Commission is not to find an alternative 22 market structure, the challenge for the Commission is going 23 to be to try to bring new participants with new trading 24 models into the existing market structure in a manner which 25 both promotes competition, promotes investor confidence in 1 the market and is fair to the other market participants. 2 MS. NAZARETH: Thank you, Paul. Bob? 3 MR. STEEL: My name is Bob Steel of Goldman Sachs 4 and I thank the Commission for the opportunity to be here 5 today and I assure Tom he's not the only one who feels 6 foolish on a regular occasion. I think a lot of what I was 7 going to say by introduction has been said so maybe I can 8 shorten the comments and say from our perspective at our firm 9 we see the equities marketplaces through multiple lenses and 10 from multiple perspectives. We act as a traditional 11 brokerage firm handling orders for clients. We are a major 12 block trading firm often using our own capital to provide 13 liquidity to institutions and issuers. We make markets in 14 over 5,000 over the counter stocks. One of our affiliates is 15 the second largest specialist on the floor of the New York 16 Stock Exchange. Another affiliate provides sophisticated 17 electronic order technology to institutions and broker- 18 dealers. We are among the largest options market makers and 19 we also own a substantial minority interest in Archipelago, 20 an alternative trading system. 21 In summary, from our vantage point decimalization, 22 technological advances in order handling execution and the 23 conversions of roles of various market participants have 24 exposed frictions in the cash and derivative markets that 25 should be addressed. Examples that have been subject to 1 recent debate include ECN access fees, the use and submission 2 of market data revenues and the effectiveness of market 3 linkages particularly the connection between auction and 4 electronic markets. Although extensive Commission 5 involvement in structured markets is often unnecessary and 6 potentially harmful due to the swift and dynamic nature of 7 change, we believe in this case it is incumbent upon the 8 Commission not only to remedy some of the existing ailments 9 in the U.S. equity markets, but also put a forward looking 10 framework in place that enables the rules and macro 11 structures of markets to evolve as the underlying 12 participants both compete and develop. Thank you. 13 MS. NAZARETH: Gunner? 14 MR. BURKHART: My name is Gunner Burkhart. I'm the 15 global head of trading at Deutsche Asset Management based in 16 London. We are an 11th hour addition to this gathering and 17 we appreciate being invited to it. We manage about $800 18 billion in assets in all the asset classes around the world 19 and our constituents are retail investors, be they in the 20 United States, Asia, Europe, as well as corporate pension 21 funds, institutional funds, et cetera. We think the timing 22 of these discussions, both in October and today and hopefully 23 the ongoing discussions, is highly relevant. We are 24 currently engaged in similar discussions with stock exchange 25 or Bourses, as they're known in Europe, and also in Europe 1 with regulators in Europe and Asia as well -- Given the 2 turbulence in our industry collectively as well as in the 3 world, there's certainly been no better any time my twenty- 4 year career to sit down with the main and major participants 5 and constituents in our industry to discuss these topics. 6 Our concerns around the world and certainly very 7 relevant here in the United States boil down into three main 8 categories and I echo some of my competitors Scott and Minder 9 this respect. They are liquidity, efficiency and 10 transparency, and I hope and I am sure that throughout the 11 remainder of the day we will enter into fairly lively 12 discussion about how we attain and regulate all three of 13 these, not just for the institutional asset managers in the 14 world, but for the investor on the street. That's not just 15 Main Street, USA but also Main Street, Frankfurt or Main 16 Street, Taipei, for that matter, who are also very large 17 investors in the United States. 18 Thank you for allowing us to participate. 19 MR. LEE: Good morning, my name is Ruben Lee and 20 I'm founder and managing director of the Oxford Finance 21 Group. May I echo everybody's previous statements that it's 22 a pleasure and honor to be here and thank you very much, and 23 may I also say that I think that these issues are very, very 24 important, notwithstanding the fact that some other issues in 25 the private markets seem to be dominating the newspapers, I 1 think these are very important and should not be eclipsed by 2 them. 3 The Oxford Finance Group is a research, consulting 4 and implementation firm and I do all of three of those 5 things. I have clients in special financial organizations, 6 government regulators, exchanges; a whole series of 7 commercial activities. I'm also an advisor to the European 8 Parliament on financial services. 9 Notwithstanding all of that, I view my role here 10 today to be to propose to the Commission to further its own 11 goals, namely, to look at the American securities markets, to 12 look at those issues which are most important, to find 13 optimal answers and most relevantly, to find answers that are 14 not so radical that they can't be considered, but yet are 15 still not so conservative that they should not be considered. 16 So in particular, I'm interested in the feasibility of 17 different types of questions, and by that I mean, the extent 18 to which different sorts of solutions may be feasible within 19 the current legal structure, so I don't think we should seek 20 to change the current legal structure, but I think that there 21 is much that is open to interpretation, that many people do 22 not yet appreciate. 23 I also am interested in the feasibility of 24 solutions, that is to say, the political feasibility, within 25 the market and I'm also interested in the extent to which 1 current regulatory structures are, in my view, in the 2 American context, captured by a range of different political 3 interests, which restrict change. We will be looking at a 4 whole series of issues to do with transparency, best 5 executions, exchange regulations, and I would be happy to 6 participate in that. Thank you again. 7 MS. NAZARETH: I think it's clear from these 8 participants we have an incredibly knowledgeable group and 9 clearly everyone comes with their own perspective, but that's 10 exactly why you're here. We're interested in carrying forth 11 with these concepts and you'll see that well over the course 12 of the day. 13 As Paul O'Kelly mentioned, we're going to briefly, 14 as we did in the last market structure hearing, discuss the 15 basic principles of the national market system and see 16 whether, again, although no one should feel compelled, 17 whether again we have a broad consensus that those core 18 principles that characterize our national market system 19 remain valid today. 20 Let me go over again briefly what they are. The 21 first is transparency of information. That is, the 22 availability to brokers, dealers and investors of information 23 with respect to quotes and trades. The next is economically 24 efficient execution of securities transactions; ease of 25 obtaining best execution; fair competition among markets and 1 intermediaries within marks and lastly, the opportunity 2 consistent with the aforementioned goals for investor orders 3 to be executed without the participation of a dealer. 4 So again I'd like to ask the group before we do go 5 on, because it's important to know whether we're all working 6 off the same principles, whether these are the right 7 precepts, and where there's conflicts, how those conflicts 8 should be resolved. It would be helpful to establish, again, 9 before we go ahead, whether there's consensus on these 10 principles. 11 This is somewhat theoretical, although obviously it 12 has very pragmatic implications for our discussion today, but 13 since it's a somewhat academic excise to begin, I thought I'd 14 begin by calling on one of our academics, it's sort of like 15 getting back at some of these academics like Professor 16 Goldschmid who called on us in Columbia, but I thought I 17 would ask Professor Kyle if you have a view that these are 18 the correct goals. 19 PROFESSOR KYLE: I think they're the correct goals, 20 except I want to think a little bit about what we mean by 21 "fair competition." To some people, fair competition means 22 if the other guy is more efficient than I am, then the other 23 guy should be punished so he and I have equal market share 24 and equal equilibrium. To others fair competition means the 25 rules of the game should be the same for everybody so the 1 person who provides the most efficient service captures the 2 greater market share. To me it means the latter, that is 3 the more efficient should capture the larger market share. 4 MS. NAZARETH: Tom? What do you think? 5 MR. PETERFFY: Well, legally, these are the correct 6 principles and we have made a lot of progress towards them in 7 the last three or four years, especially decimalization was a 8 very large step. These objectives are completely within our 9 reach, if only we could just get along. 10 With technology and sufficient economic incentives, 11 I think that the participants and the Commission can achieve 12 them in a very short period of time. 13 MS. NAZARETH: Bob Steel? 14 MR. STEEL: I was just laughing. I'm sure "get 15 along" in Tom's vocabulary means we do it his way. 16 MR. PETERFFY: Or yours. 17 MR. STEEL: I've got lots of ways. 18 You know, Annette, I think that it's probably 19 pretty easy to pledge allegiance to these ideals, but really 20 the challenge is the nuance at the edges and we heard a 21 dichotomy, I think on this side of the table, about 22 transparency is a good ideal, but at some point too much 23 transparency carries with it a burden that people who are 24 representing large numbers of individuals with a larger order 25 feel disadvantaged and so I think the real issue is threading 1 the needle on these three or four issues with greater 2 specificity and we can all kind of, as I said, pledge 3 allegiance to the ideals, but then it gets down to really 4 determining how you get down to this dynamic of Bob's 5 complete transparency and Scott and Minder's idea of 6 transparency which has a different idea and cost. I think 7 this is the tricky part, especially in the area of 8 transparency. 9 MS. NAZARETH: You alluded to this also, Ruben that 10 to some extent it's in the eye of the beholder and people's 11 obviously own business models or political inclinations can 12 factor into this. Do you have more comments on this? 13 MR. LEE: I have a couple of comments; one on 14 transparency, one on fair competition. On transparency, I 15 would make three comments. The first is that what you're 16 getting now is not what you think you're getting, so you 17 think that you're getting the best bid and offer, but you're 18 not actually getting that, but most people don't appreciate 19 that, I think in the American context retail. The second is 20 that I very much agree and I think there's a fair amount of 21 academic evidence to support Minder's viewpoint that optimal 22 transparency is not full transparency. There are a range of 23 different tradeoffs, but in my mind, I'm very much on that 24 point, and we can go into the subtleties of that. 25 The third is and this is going back to the '75 1 amendments which is very important, is that, Pete alluded to 2 this, which is that market structures of many different types 3 have strong incentives to provide transparency, number one. 4 The second is that even where there are no mandated 5 links between markets, one can point to Europe, we have seen 6 consolidators of information, so the idea that we need to 7 have some sort of institutional consolidator of information 8 here in the U.S. I think is much less relevant than it was 9 when it was initiated and in part because it's so much 10 cheaper to do that. So I think one needs to be careful about 11 requiring a need for mandatory transparency when there's so 12 much private incentive to produce appropriate levels. 13 Of course there are exceptions to that, where 14 transparency must be structured and mandated in some way, but 15 I view that to be an exception. 16 Let me now make a comment on the fair competition. 17 In my view, in most contexts, there are very strong 18 incentives for orders to centralize on particular trading 19 systems and in fact what we see here in America is actually 20 very unusual, that we have such an abundance of trading 21 systems, notwithstanding the fact that they are shrinking in 22 number and in part I think that abundance does arise as a 23 result of various SEC rules which allowed the continuation of 24 a range of different systems which in fact are being 25 subsidized in various ways by these rules. 1 I think there's a counter balance to that, which is 2 if you think as I do that orders may tend to consolidate that 3 where you have institutions such as the New York Stock 4 Exchange where they do consolidate, I think there are very 5 good grounds for examining the actions of such monopolistic 6 providers very, very carefully and a whole range of different 7 aspects and while in many contexts it is appropriate to say, 8 yes, leave people to decide what sort of market structure 9 they want to adopt and I agree with that, once we start 10 approaching the monopolistic provider, I think there we need 11 to be much more careful. 12 MS. GLASSMAN: I have a question for anybody. 13 Given, there seems to be general agreement on what the ideals 14 are, perhaps not the specifics, can we meet all of these 15 ideals simultaneously, are there tradeoffs we have to make 16 among them? 17 MR. CHENG: Maybe can I start? I think the 18 question, maybe if I can rephrase it, is from the 19 Commission's point of view, are you in a position to design 20 or promote a market structure that basically is one size fits 21 all. The question we have here is the market is full of 22 retail and institutional investors. Going back I think to 23 the five principles behind the NMS, the third being the ease 24 of obtaining best execution, we already heard earlier that 25 best execution means different things to different people. 1 To the retail investors it may mean the best price, 2 but for institutional investors, it may not mean the best 3 price at the moment of trade, because again, going back to 4 the fact that we often trade large blocks over time, so the 5 question is, how do we create the right balance between the 6 two? And back to the transparency. I think we can achieve 7 it, although the question in our current market, I think 8 Scott mentioned as well, that transparency cannot be achieved 9 without protection, because if there's no protection, in 10 other words, there's no, say, price, time, size priority, 11 then I think the market participants will not be incentivized 12 to really show their hands so they could get picked up. 13 So I think on the one hand we are basically dealing 14 with competing goals here. The five elements are all good on 15 paper. The question is, can we somehow find a way to piece 16 them together without conflict. Thank you. 17 MR. DeSANO: I just wanted to add, I don't believe 18 there's a perfect market structure, period. And that we need 19 to think about that as we think about all these things. 20 There's a little give-up on everybody's part because of all 21 the constituencies. Competition and fairness lends itself to 22 continuous improvement in the market structures, but we'll 23 never get to a perfect structure. 24 MR. DeSALVO: I think what would help some of these 25 issues that we're talking about, would be from our point of 1 view a clear definition of best execution. Fragmentation and 2 competition has drawn away the retail orders into shades of 3 gray with regards to best execution away from price. Market 4 accessibility between, where we have auto ex in certain 5 marketplaces and not other marketplaces have led even retail 6 order flow to be subject to shades of gray when it should 7 clearly be price subject to best execution at a certain 8 level. We don't even have that in this case and what we 9 would like to see coming out of this is a clear definition of 10 best execution with getting back to price and the only way 11 that you can get back to price with regards to best execution 12 is markets that efficiently transact between each other as 13 opposed to the current market structure where there is 14 inefficient linkages and inefficient ways that markets access 15 each other. 16 MR. DOMOWITZ: Getting at the original question, 17 which is whether or not you can satisfy all these at once, I 18 mean, there's a small preamble to that. We can all agree, 19 I'm with Bob, it's easy to pledge allegiance. This is like 20 motherhood. Frankly, it's easy to agree. Following up on 21 Ruben, if we didn't agree, we'd have to change legislation. 22 This is not something we can actually agree on at this table 23 by ourselves in some sense, but it's worth saying, first of 24 all, market structure is a term used by a variety of people 25 to support a variety of positions. It is essentially in my 1 view, the rules and institutions that contribute to 2 competition in the marketplace for transaction services. 3 Now, viewed that way, what's feasible now, all 4 right, again, following up on something Ruben said, is indeed 5 I think the reconciliation of these five points. Should we 6 indeed get a little bit deeper to these rules that are 7 promoting competition in the first place and to me that is 8 the right place to start. 9 Absent that, were I to go back to the legislation, 10 I would actually turn to this point about best execution and 11 I would say let us create best execution obligations and the 12 rest will follow. 13 MS. NAZARETH: Janet? 14 MS. ANGSTADT: I was going to say, I think all 15 these questions go to best execution. I would urge that we 16 not try to define best execution any more than we have. I 17 think the 1-5 rules have really offered the disclosure that's 18 necessary for people to make their choices. Again, too, I 19 think it was Minder's point that best execution to an 20 institution may be very different than a retail investor and 21 perhaps some of the suitability rules that we have are 22 already in place to help protect retail investors and I think 23 the diversity that we have in the market structure today, 24 really allows people to choose speed over price if that's 25 what's important, so I think actually these principles help 1 lead us to that kind of a best execution very strongly. 2 MR. DeSALVO: I hate to go back and repeat this, 3 but again, we don't even have a situation right now where we 4 have an agreement on smaller, for the retail institutions, of 5 what is best execution, because they're opposing views of 6 speed or price. I think if the Commission can come out and 7 say what is the definitive goal, at least that would allow 8 all of us to be on the same playing field rather than the 9 current shade of gray. 10 MS. NAZARETH: Well, I think, as you know, Matt, 11 that the analysis is so colored by other principal/agency 12 conflicts, like payment for order flow and the like, that 13 it's not as simple as just speed or price. But we'll get 14 into that later. 15 Does anybody else have any points before we take a 16 break? Yes, Paul. 17 MR. O'KELLY: I'd like to make one point. While 18 these principles are very noble, if they suffer from 19 anything, they suffer from a little vagueness. But if we 20 look at the '75 Act amendments where these principles were 21 articulated, you'll see in the very next section where 22 Congress makes the point that markets that are linked 23 together can promote these five principles. Now, how we 24 interact with each other is a very difficult issue to 25 address, but it seems clear to me that Congress back in 1975 1 didn't like the idea of market fragmentation, didn't think 2 that it would promote these five principles and I have no 3 reason to believe that that's changed in the subsequent 27 4 years. 5 MS. NAZARETH: Thank you. Yes? 6 MR. MURPHY: I'd just like to, a couple of points. 7 As far as best execution, I think as long as the ultimate 8 customer is at the forefront and his outlook is what's 9 paramount, the markets in general will acquiesce and should 10 be encouraged to just keep the ultimate customer primary, 11 primary concern. 12 When it basically gets to a point where best 13 execution is defined by economic terms or being able to take 14 advantage of a situation, that's where we get to that gray 15 area that Matt talked about, and right now, the 1-5 data that 16 we've seen so far I think is pretty skewed and it's not 17 talking about apples and oranges. I think we can do a lot 18 better or maybe get some more independent input on best 19 execution statistics, so you know, I basically agree with all 20 these principles, but giving the customer choices, because 21 definitely, there are different venues that benefit the 22 different participants in the markets, so as long as there 23 are choices and they're transparent choices, we will do okay. 24 MS. NAZARETH: Yes? 25 MR. LEE: Yes, I've got one more point. To be very 1 explicit, I think that there are tradeoffs between these five 2 different goals, and I think that the choices that the 3 Commission has made have implicitly showed their choices for 4 these different goals. So in particular, if you look at the 5 fifth one, which is investor access, subject to the first 6 four, one could debate whether that one has been promoted 7 given membership structures of exchanges and so on and so 8 forth, and in fact one could strongly argue that it has not 9 been promoted, so these goals have not been equal in the eyes 10 of the Commission historically. 11 MS. NAZARETH: It sounds like yet again we have 12 violent agreement on the principles, and we'll spend the rest 13 of the day arguing over the details. 14 We'll take a fifteen-minute break and reconvene at 15 five of eleven. 16 (Brief recess.) 17 MR. COLBY: Welcome back, everyone. We're going to 18 turn to the issue of market data as one of the fundamental 19 building blocks. I want to make sure everybody talks into 20 the mike, because the audience has had trouble hearing. 21 We also have two players replace people who were 22 injured in the prior panel. Duncan Neiderauer is replacing 23 Bob Steel. Actually, he wanted to do it, but Duncan pushed 24 him out of the way. 25 U.S. investors today have access to a consolidated 1 real-time stream of market information for each of the 2 thousands of stocks traded in the U.S. markets. This price 3 transparency is a cornerstone of the U.S. national market 4 system. It facilitates the best execution of customer 5 orders, promotes investor protection and mitigates the 6 fragmentation of buying and selling interest among different 7 market centers. The most basic form of market information is 8 the best quotation and last sale data with respect to a 9 particular security at a given time. The best quotation is 10 the highest bid and lowest offer price for a security 11 currently made available by self-regulatory organization and 12 is commonly referred to as a national best bid and offer or 13 the NBBO. Last sale data generally identifies the price at 14 which the most recent trade in a particular security 15 occurred, the size of that trade, and the market in which the 16 trade took place. 17 Price transparency in the United States, as we 18 discussed earlier, was mandated by Congress in 1975. 19 Congress directed the SEC to use its new authority to adopt 20 rules to integrate the market through the collection and 21 consolidation of market information. Under these Commission 22 rules, quotes, trade price and volumes are collected by the 23 stock exchanges and NASD for all exchange listed and NASDAQ 24 stocks within seconds of the quote or the trade taking place. 25 The current NBBO quotation, as well as individual market 1 center, over the counter market maker and ECN quotes and 2 reports of all trades are disseminated for each stock on a 3 nearly real-time basis. Vendors and brokers that make this 4 market information available from--vendors that make it 5 available from any individual market are required by 6 Commission rules to make equally available consolidated 7 information from all participating markets. 8 Furthermore, to implement this quote collection and 9 trade reporting mechanism, the SRO's, the self-regulatory 10 organizations, have acted jointly under several national 11 market system plans to disseminate consolidated market 12 information. These plans govern all aspects of the 13 arrangements for collecting and distributing market 14 information. Among other things, they require the individual 15 SRO's to transmit market information to a central processor, 16 which then consolidates the information into a single stream 17 for dissemination of vendors and some large end users. In 18 turn, the vendors, brokers, and large end users may 19 disseminate the information to the public. 20 These plans also establish fees for market data. 21 The resulting net revenues are divided among the 22 self-regulatory organizations that provide the data and 23 represent a significant percentage of the SRO's revenue. The 24 plans also establish non-fee standards such as reporting 25 obligations, hours of operation and regulatory halt 1 procedures. 2 The implementation of decimal pricing and point 3 increments in 2001 has increased the demand for deeper or 4 different levels of market information, as was touched on 5 earlier in the discussion. For example, the New York Stock 6 Exchange recently began to provide real-time semi-dynamically 7 updated view of the limit order books for all New York Stock 8 Exchange traded issues. NASDAQ Super Montage, which began to 9 operate on October 14th, provides NASDAQ stand alone 10 quotation and ordering system by a means of aggregating 11 quotes at a multiple of price levels to create an integrated 12 trading system. Some ECNs such as Island and Archipelago 13 make their limit order books available on their websites at 14 no charge and some Internet portals also make this 15 information available for free for just users on the web. 16 Nonetheless, despite these steps taken by the 17 markets, the impact of decimalization led some market 18 participants, as was evidenced earlier in the discussion, to 19 believe that more limit order information is needed in 20 standard formats that can be integrated. There have also 21 been repeated questions raised regarding the market data fees 22 that are charged by the consolidated systems disseminators. 23 In particular, the fact that market centers are rebating or 24 seeking to rebate a significant percentage of these market 25 data revenues to their members, has led some to suggest the 1 market data fees are too high, and could be reduced 2 substantially. 3 In light of the wide range of views and comments 4 that are received on a concept release that was put out by 5 the Commission on market data revenues, the SEC formed an 6 advisory committee in August, 2000, to assist it in 7 evaluating issues relating to market value data available. 8 In a final report in September of 2001, the committee 9 reaffirmed the importance of price transparency and 10 consolidated information, the importance of these for the 11 health and vitality of the securities markets and then it 12 recommended with varying degrees of consensus a number of 13 measures to increase flexibility in market centers and data 14 vendors to make data available to the public. 15 It also suggested that technological and 16 competitive developments may have lessened the need for 17 market centers to act jointly to consolidate data under 18 existing national market system plans. 19 So the purpose of this session is to consider the 20 ways that market data, in particular real-time market data, 21 is collected, processed, and made available and the many 22 questions that have arisen regarding this data. 23 These include the utility of the data for 24 investors, and we need to talk about for which investors, the 25 way that the data is made available, the fees that are 1 charged and inherent in talking about the fees that are 2 charged, I think is the topic of how self-regulatory 3 organizations get funded, even though that's a major issue 4 that comes up later in the day. 5 So, with that information and introduction, let's 6 start in on the questions. I wanted to start first in 7 talking about the information that's available, is it useful, 8 what is its nature, is it useful, then go to should other 9 information be available and then after we have talked about 10 that, start talking about fees, if that's all right with the 11 panel. Fees are going to be an interesting topic in and of 12 itself. Let's start with the national best bid or offer. 13 The national best bid or offer is in a sense a product built 14 and maintained for retail investors, and so I think, first 15 question should be how useful is it? How important is it to 16 have some sort of a measure of the existing quote and how 17 good a measure is the NBBO? Anyone want to address this from 18 a retail investor standpoint? Ruben, you expressed some 19 views about the nature and utility of the NBBO. Do you want 20 to start? 21 MR. LEE: I think several points. The first is, I 22 agree with Pete's point, that if you are relatively 23 uninformed, that as much needs to be done, that should be a 24 focus of the Commission and it clearly is not, given your 25 historical mandate to investors out there. So retail 1 investors should be a primary concern. 2 The best bid and offer does play a very, very 3 important role in showing what may be available, but I think 4 that it's, as I indicated beforehand, it doesn't show the 5 full bids and offers that are available throughout the 6 market, and there's a question in my mind, both as to whether 7 it could do that or whether it should do that. That's the 8 first point I'd like to make. 9 The second is that, and this comes to the issue of 10 competition between different markets, and both as trading 11 institutions and in terms of their ability to gain revenues 12 from data. 13 If you considered the value of data from competing 14 markets, you would probably say that the New York Stock 15 Exchange data is much, much more valuable than anybody 16 else's, and on that basis-- 17 MR. COLBY: You're speaking about quotes or trades 18 as well? 19 MR. LEE: Let's talk about quotes initially. 20 On that basis, there's a question as to whether the 21 SEC's rule, which says that if you carry any particular data, 22 you need to carry all data, okay, is necessarily appropriate, 23 indeed, whether it follows from the Act, that's to say, you 24 might choose, even if you decided you accepted that 25 transparency was an appropriate goal because you're mandated 1 to, whether you needed the rule to deliver that goal, and I 2 think there are questions about that, which lead to questions 3 specifically about the options market, where, and I know that 4 you're familiar with this, where there's an excess of data 5 with all these different dates and so on and so forth. There 6 the vendors are complaining that you may mandate it because 7 you have a regulatory interest in getting that data, but 8 actually the cost of getting all that piping needs to be 9 borne in mind. 10 MR. COLBY: Several issues raised and they're all 11 important. Pete, did you want to address this NBBO issue? 12 PROFESSOR KYLE: If you go back to the five points 13 from the previous session; transparency, economic access to 14 the market, best execution, fair competition and facilitating 15 investors order crossing, it seems like having a national 16 best bid best offer is necessary to achieve at least four of 17 those, because the basic idea is that you want a small 18 investor who closes an order with one securities firm to buy 19 at 4, another investor places an order in another securities 20 firm to sell at 3, or sell three or four. A transaction 21 ought to occur and at a very minimum in order to achieve four 22 of those objectives, you need some mechanism for keeping 23 track of the fact that a transaction ought to be occurring, 24 and that requires having national best bid/best offer. 25 Now, if you want it to be set up in such a way that 1 an economically relevant sized transaction can also be 2 monitored, then you want to show a certain amount of depth to 3 transactions, because it's possible the best bid might be for 4 a tiny quantity. Even for a retail investor it might be 5 relatively smaller than his order so you want to show at 6 least enough depth to show a reasonable sized quantity, be 7 able to be monitored with respect to the prices offered in 8 the market. So it seems to be a requirement that that be the 9 case and probably a requirement that it be augmented enough 10 to have relevant sized quantities in it. 11 MR. COLBY: Bob, would you like to talk about what 12 the New York Stock Exchange is doing to respond to that? 13 MR. MURPHY: Sure, just how I feel about it, and 14 again, there is a little bit of a split at the stock 15 exchange, but I do feel that there is a need for the NBBO, 16 and it's -- definitely in a decimal environment it's been 17 diminished, but, you know, the issue of not needing it any 18 more, it's almost we're going too far the other way. We 19 would be taking transparency out of the market at the expense 20 of the little guy, so I think we definitely need it. 21 But depth, as the Professor said, is paramount. Now 22 because of decimals and we have our Open Book, that's been 23 very well received as a data product. I think we have over 24 7500 subscribers now since it's been launched, and we also 25 have, I believe before the Commission, Dual Quote, which will 1 include NBBO, include the Open Book, and will also force the 2 specialist to put up a size held market that all participants 3 can interact with as need be. So-- 4 MR. COLBY: Meaning that it's a market that's below 5 the NBBO or above the NBBO where there's a greater size 6 available? 7 MR. MURPHY: Absolutely, and should and must be 8 disseminated to all participants. So you know, that's our 9 answer to the lack of transparency in a decimal environment 10 and I think the more--that also, it includes the crowd, it 11 includes all orders entrusted to the specialist and the 12 specialist's interests as well, to give institutions a better 13 look at what kind of liquidity is down there. 14 MR. COLBY: Matt, you have trading prices and a 15 large retail brokerage operation. What role does the NBBO 16 play? 17 MR. DeSALVO: I think the NBBO has been diluted 18 with the introduction of decimals. It had much more 19 importance at the steep level than it does at the penny 20 level. The introduction of Super Montage and Dual Quotes and 21 Open Book I think are representations that the marketplace 22 recognizes that the NBBO needs to add some depth, so you've 23 had exchanges respond to it. I think the important thing 24 from our point of view is that the NBBO is the basis point 25 with which you start making some of your decisions, but 1 accessing the NBBO I think is where we come out on whether it 2 is important or not. 3 In other words, an NBBO that is represented by a 4 market that is easily and fairly accessible is meaningful. 5 An NBBO that is representative of underlying markets that are 6 difficult to get to, slow in response time, et cetera, where 7 you start to grade away from an auto-ex environment, leads to 8 NBBO's that are different. An NBBO in one stock versus an 9 NBBO in another stock with equal one penny bid offer spreads 10 can have entirely different meanings with regards to response 11 time and accessibility, so I think that a marketplace that 12 has an NBBO as a minimum requirement for market data 13 dissemination you have to also step back, the NBBO goes up in 14 importance if there's equal accessibility between markets. 15 MR. COLBY: So the accessibility is a key question 16 with the NBBO. We're going to leave actual detailed 17 questions about access to the later session. But I note the 18 point if you have an NBBO, and it can't be accessed, it's a 19 very difficult question. 20 MR. COLBY: Could I ask Minder what role does the 21 NBBO play and what do you look at, I'm going to ask Gunner to 22 address the same thing. 23 MR. BURKHART: I think we probably have the same 24 answer, I wouldn't want to speak for Minder, but the NBBO to 25 us is nothing more than a guideline or an indication. Matt's 1 point is well taken, in that it's the depth and the 2 accessibility that's the real issue for institutional 3 investors. You have to have an NBBO I think for the retail 4 world out there, and even for us to have a guideline that 5 says that even if it's a hundred share market on the bid and 6 the offer, that's the best bid and best offer out there. 7 It's really the depth beyond that and the accessibility, 8 whether it's a hundred share market or the hundred thousand 9 share market which Bob is referring to, the capital 10 commitment market outside of that, the specialist is now 11 willing to put up, it's of more interest to us. 12 Yes, the NBBO is an essential item. However, it's 13 largely irrelevant for an institutional investor. Sorry, 14 Minder, I didn't want to speak for you. 15 MR. CHENG: I think we're in competition across the 16 table but I think our views pretty often are similar so no 17 further comment. 18 MR. LEE: Could I make one comment about the NBBO? 19 I'm wondering what is the incentive for markets not to 20 provide their quotes. If there are incentives not to provide 21 their quotes, to what extent do you guys need to be involved? 22 That's what I'm wondering. Not to put the questions to the 23 listeners today, but-- 24 MR. COLBY: I would just say that we went through 25 an era where we needed to mandate quotes to be made available 1 because there are mixed incentives from different trading 2 systems. It's worth, can I pick up on one of your--unless 3 someone wants to rise to that, could I pick up on one of the 4 questions that you posed, Ruben, which is that we currently 5 have at SEC a rule which says that if vendors or brokers give 6 information about one individual market to customers, they 7 must provide equally the information from all markets that 8 are participants in the NBBO. There were a number of 9 historical reasons for that. You posed the question whether 10 it's still necessary. Recognizing that some markets have the 11 larger percentage of the best quote than others. 12 Does anyone, Paul, is this something you would be 13 interested in talking about? 14 MR. O'KELLY: Well, the requirement for the Vendor 15 Display Rule is one that I think it's important from public 16 investor protection reasons, because if you do not require 17 all folks to be part of a consolidated quote and vendor 18 display rule, some market participants will only buy a New 19 York quote or only buy the NASDAQ quote. And that's going to 20 have a spiraling effect over time. If fewer people buy other 21 market competitors quotes, market competitors are going to be 22 less able to provide markets that compete with the markets 23 who are selling their quotes and eventually you're going to 24 lead to one, a single market selling its quotes and there 25 will be no alternatives. 1 If you do not allow other markets to display their 2 quotes, even a consolidated quote, to give customers the 3 opportunity to choose among them, you're going eventually 4 lead to a single both on the listed side and NASDAQ side. I 5 don't believe people are anxious to have that. 6 MR. COLBY: So there's a limited ability to get 7 your quotes disseminated. I'm done. 8 MR. O'KELLY: You earlier talked about tradeoffs in 9 these principles that Annette articulated and in this NBBO 10 area, I think this is one of the areas where a trade-off 11 occurs, is that you have the NBBO where people have to act 12 jointly, because we believe price transparency is a good 13 thing, it will make everyone join together to provide 14 transparency to everyone for the NBBO. Beyond the NBBO there 15 can be a relaxation of that requirement and allow markets to 16 compete with each other. If the New York Stock Exchange 17 wants to provide its open order book to its customers because 18 they believe it will give them a competitive edge, they 19 should be free to do so. If NASDAQ wants the Super Montage 20 to show the market three quotes deep, they should be free to 21 do so. 22 This is an area where I think the Commission has 23 acted responsibly in allowing the goals of transparency and 24 competition to co-exist, requiring vendor display, 25 consolidated NBBO, on the other hand, allowing markets to act 1 on their own competitively to sell products beyond the NBBO. 2 MR. LEE: Can I take issue with that, in the 3 following sense, that if I'm an investor, clearly I want to 4 know where I can execute my trades. Because I want to know, 5 someone's going to fulfill that demand for me. A vendor is 6 going to package together exchange X, exchange Y and exchange 7 Z. Suppose that the vendor decides that, or rather, even the 8 investor decides that actually exchange Z is not really worth 9 knowing about because I'm not going to undertake much 10 business on that. Nobody's saying that they should not be 11 allowed to package the data together, but the question is 12 should any be required to package the data together and 13 indeed if they are required to package the data together, 14 does that have implications of supporting inefficient trading 15 systems. That's where I think you might want to think about 16 it. 17 MR. GOLDSCHMID: I want to get a question in. The 18 problem of entry barrier in your analysis. 19 MR. LEE: Entry barrier is a real problem in which 20 there are market externalities, no question and it may be 21 that you want to effectively subsidize entry barrier, which 22 is what is going on here, but on the other hand, by doing 23 that, it's important to recognize that you are subsidizing 24 it, and that's what I'm saying it. 25 MR. COLBY: To what extent also is there an 1 agency-principal problem, because, it sounds like what you're 2 setting out is a model for sophisticated investors choice of 3 information, but to the extent this is as, people said a bare 4 minimum that's used for retail investors, to what extent do 5 you think that the choice would be in the hands of the 6 investor as opposed to the purveyor of the information to the 7 investor? 8 MR. LEE: People go to Tom Gardner and he says he's 9 a fool, but he's not a fool, right? And they go to him 10 because he's built up a reputation, and the issue is, if I'm 11 a vendor and I'm going to be offering something which retail 12 investors who know that they're relatively unsophisticated, 13 but can trust him, that is something that I would want to 14 invest in, so it doesn't seem to me, there's still a private 15 incentive to be promoting. 16 MR. COLBY: Let's turn to another issue that was a 17 very big part of the advisory committee consideration of 18 market data. Let me just put it out, whether it's of any 19 interest to the people in this room, and that is the question 20 of how the data gets pulled together. Is it important to 21 have a central consolidator that pulls the data together, or 22 is it sufficient to require information to be made available 23 and either leave it open to people to pull it together or to 24 say it needs to be consolidated without a central 25 consolidator? My first question, is that of anything that's 1 of any real interest to anyone at this table and if so, what 2 are your views on it? 3 It's a real interest to many market participants, 4 Ruben, I'm going to hold you for a minute, because we see all 5 these things are of interest to you. Anyone else? 6 MR. DeSALVO: I would say with regard to SIP's 7 work, whether it's one SIP or multiple SIPs competing-- 8 MR. COLBY: By SIP you mean a central processor? 9 MR. DeSALVO: Yes. The actual process of 10 consolidating disseminating is not broken. What is broken 11 and we hope that you get to it, is the cost that we pay for 12 it. 13 PROFESSOR KYLE: I have some views on this. 14 MR. COLBY: Pull the mike up. 15 PROFESSOR KYLE: I think that having consolidated 16 data going out through one feed has potential advantages, 17 especially if you augment it to put more audit trail type 18 data in it, because then you automatically get a fair time 19 sequence, assuming all the different endings that are feeding 20 data into it are feeding data into it in approximately the 21 same way and what comes out at the end is going to have a 22 certain kind of time sequence that's very integral to 23 monitoring whether customer orders that are sitting there in 24 the market are respected or not respected, there's a one- 25 second difference in execution time. 1 MR. BURKHART: I would say also that in respect to 2 the principle or ideal of fairness, if you like, in 3 transparency, if you have a consolidated data source at least 4 you know or you could feel somewhat comfortable that all the 5 participants, be they retail, institutional or otherwise, 6 with brokerage are receiving the same level of data, and what 7 they do with it and how they handle it, or how they repackage 8 it and analyze it to their customers or internally is really 9 where their efficiencies or competitive advantages should be 10 borne, as opposed to who receives what. 11 If we all received the same data from a single 12 source -- if Matt at Morgan Stanley gets the same data as 13 Duncan at Goldman Sachs, what they do with it will determine 14 their competitive edge over one another. Ditto if Minder and 15 I received or Scott and I received the same data. How we 16 analyze it, restructure it, retool it for our own purposes 17 really should be the different shading factor amongst us, not 18 multiple data feeds that we have to pay different amounts 19 for. 20 MS. ANGSTADT: It seems, too, that it becomes such 21 a tool of the regulators to, in the examination for best 22 execution and it's really been a measurement of best 23 execution that, even if somebody wants to buy unconsolidated 24 data, they really do need for regulatory purposes the 25 benchmark of the NBBO just to measure where they are. I 1 think one aspect of the Vendor Display Rule that I don't know 2 if we'll get into is the display of the last sale data, what 3 evil will befall us if we're able to display on a real-time 4 basis and unconsolidated basis the last sale as opposed to 5 just the NBBO. 6 MR. COLBY: What evil will befall us? 7 MS. ANGSTADT: I don't know that any will. Is 8 there a reason that we should not, you know, revisit the 9 Vendor Display Rule and relax some of those requirements so 10 that while we've been able to provide real-time limit order 11 books and again on an unconsolidated basis so people can take 12 a look at this information, what about providing the last 13 sale as well, if that's something that a marketplace chooses 14 to do. 15 MR. COLBY: Since not everyone has memorized the 16 display rule, what it requires is that if you make the quotes 17 from any one market available, you have to make the quotes 18 for all markets, and if you make the last sale for any one 19 market available, you have to make the last sale for all on 20 an equally accessible manner, but it doesn't require that you 21 make available the quotes of any market, it doesn't link the 22 two, the quotes and the trades together. So, Ruben, I know 23 you want to talk on these two issues. 24 MR. LEE: A couple of points. The first is, to use 25 an example in Europe, we have competing markets and private 1 market participants, either vendors have consolidated without 2 any mandatory requirement to do so. That's the first. The 3 second is I would take issue with two things which Gunner 4 said as follows: I think that his notion that if we have it 5 all together in a single consolidated fee, we all work from 6 the same information base, is wrong. I think that in fact, 7 market participants do work from very, very different 8 information bases. They see all different sorts of order 9 flows. In fact, not only is it wrong, but it gives the wrong 10 impression in that having a BB, a best bid and offer makes it 11 seem as though we all have the same information, but in fact 12 we do not. 13 And the second point I would make, which is a 14 classic economist dilemma, thank goodness I'm no longer an 15 economist, which is by requiring everyone to use the same 16 structure you destroy incentives to provide better and faster 17 structures and in fact, that is part of what makes and 18 enhances the market, so I think that it gives the wrong 19 incentive structure. 20 MR. BURKHART: Perhaps our disagreement on this 21 point is founded over our definition of the data that's 22 provided. I was referring really to the basic standard trade 23 data that should be available in a consolidated fashion to 24 all participants. Obviously, each of the participants, 25 whether it's at the brokerage end or at our end, have lots of 1 different internal information that we have gathered from our 2 own proprietary sources which may or may not give us an 3 advantage, and I wouldn't want to imply that we should either 4 make that data wholly available to everybody, thereby giving 5 up our proprietary advantage, or in fact that would be of 6 interest to anybody other than ourselves. I think our point 7 perhaps differs on the data itself more than anything else. 8 MR. COLBY: Maybe we should talk--a number of the 9 panel participants said that after decimals, that there was a 10 crying need for more and different information for trading 11 purposes. It would be worth talking, and I know that Gunner 12 and Minder took it to another level in terms of what the 13 terms on which it should be available would be, but what is 14 needed, sort of public information at this stage. Tom, you 15 represent some very active trading. What are your thoughts 16 on this? 17 MR. PETERFFY: We pull the data together ourselves. 18 If we did not do that, we couldn't provide best execution. 19 Our problem is that that data means different things to 20 different people, because some of it is electronically 21 accessible and some of it is not. Now, what do you do when 22 you see a quote that is a penny apart, and the one that's a 23 penny better is not accessible. This is the problem. So I 24 think that the market data question basically depends on 25 market structure. Once we have a market where all quotes are 1 electronically accessible, the best solution automatically 2 falls out. 3 MR. DOMOWITZ: I think the market data question is 4 not just a question of market structure, it's a question 5 from, let's say, my perspective of also quality service. 6 Now, in the same way as Tom mentioned, we also pull the data 7 together. I find it very interesting a discussion of what 8 you might call size adjusted quotes, okay, size adjusted 9 spreads, you like to know what's there at a penny, what's 10 there at a nickel. When we talk like that, what we're really 11 just saying is we'd like to see some notion of an order book, 12 that's all that really is, that's basically saying there's a 13 certain amount of shares available at certain prices as you 14 step down the price panel. Now, we put that type of data 15 together ourselves for the benefit of our clients. In other 16 words, we pull data from a variety of sources, we use a 17 variety of vendors to do that, and, frankly, we don't even 18 find that in and of itself completely sufficient to meet what 19 we believe are the needs for execution. 20 We also need, if you like, scorecards for how well 21 the markets are doing. So when we talk about what do you 22 actually need to get the job done, it's actually not even 23 restricted to prices and clients, we actually need a lot of 24 things. But we can do it, we're pretty happy that the market 25 does it, we happen to think that the market does it rather 1 widely right now. It's not a question of having to mandate 2 the notion for this quasi order book for this information 3 that's provided to all participants, the market is basically 4 providing this and with means to access, although I would 5 agree with some of the comments around that access is often a 6 trifle difficult. 7 MR. COLBY: The limit order book display, other 8 than the top, is a voluntary market driven system and some 9 systems like Archipelago and Island makes its entire book, 10 New York is making its book available, NASDAQ is working 11 towards. Is this producing what the market traders need and 12 is it having a negative effect on reducing people's 13 willingness to put information on the book, and Scott, maybe 14 I can, you said I think earlier that post pennies that you're 15 doing a lot more trading upstairs and on ECNs. Is that 16 trading visible or is it less visible than it was before or 17 what's the thinking behind that? 18 MR. DeSANO: The reason we go upstairs is to take 19 some of the problems of market structure off of our plate and 20 put it on the Street, to be honest. It's an unpleasant thing 21 for them. It's more motivated by splitting trades on the 22 Floor that we don't want to split. It has nothing to do with 23 really what we're talking about here. 24 MR. COLBY: Minder? 25 MR. CHENG: I think just one more comment. I think 1 taking a step back, what is the reason for NBBO? Before even 2 getting to that, I think Gunner mentioned, for the buy side, 3 it used to be, when the buy side didn't really have much 4 direct access to the market, when we traded just about 5 everything through the broker-dealers, so which means that 6 even if I could see there was a better quote on this exchange 7 as opposed to the other exchange, the relevance was very 8 little, because we still traded through a broker-dealer, so 9 we basically imposed that responsibility on the broker- 10 dealers. Not these days. These days with this ongoing 11 intermediation from the sell side and also with the 12 technology that offers the sell side the ability to directly 13 access the market. 14 Now, here comes the question that I raised earlier. 15 If we put, what we are trying to do is basically trying to 16 figure out the order book or supply/demand. But post decimal 17 I think lack of protection, as Scott put it, basically, it 18 has forced the buy side in a way to trade more upstairs and 19 that is indeed true. And, so, in other words, there's an 20 order book, there's an NBBO and we can even address the 21 depths, but there's also that liquidity that are not showing. 22 We often go through ATSs and ECNs and we place our order 23 through the reserve book. We don't show the sizes of the 24 market because, again, because of the lack of protection by 25 showing it we will get picked off, most likely. So I think, 1 again, going back to Ian's point, the data structure or 2 whether we want to create the right data that, again, that 3 would sort of fit the retail need and the institutional need 4 for them, meanwhile, there's no protection that has been 5 offered to guarantee that whoever is providing that 6 transparency to the market can be protected and rewarded. 7 So I think that as long as we can't, I mean, to the 8 extent that we cannot resolve that issue, how much do you 9 want to provide through NBBO, how much depth, I think will be 10 ongoing. 11 MR. COLBY: Can we follow this up? Earlier I think 12 Minder and Gunner both said they need to be rewarded for 13 putting limit orders on the book with some kind of a price- 14 time. 15 MR. BURKHART: I don't know if I would use the word 16 "rewarded " in front of the regulators. I certainly am not 17 looking for any financial reward. 18 MR. COLBY: Incentive? Well, you're probably the 19 only person. Incentive for putting the orders. Could we 20 proceed on this for just a while? How much price/time 21 priority would it require to encourage institutions to put, 22 and ultimately, is there any amount that's going to encourage 23 you to put the full size of a major order on the book or is 24 that something that has to be worked, that ultimately has to 25 be worked upstairs? 1 MR. DeSANO: It's not just price/time priority, 2 it's I don't want to be improved. If in order for me to put 3 an order on the book, I want to reach two or three cents and 4 take that stock from that person who's provided that 5 liquidity on the book, I don't want to be improved. I don't 6 want somebody to step ahead and sell it to me a penny 7 cheaper, I don't want it. If I get that, I'm not putting an 8 order on the book myself, all I'm providing is a way to get 9 picked off. So it's beyond price/time priority. Does that 10 make sense to you? 11 MR. COLBY: Are you saying when there's an order in 12 the market you want to be able to access it? 13 MR. DeSANO: I want to be able to access it, okay, 14 but in order to make it work, in order to get that limit 15 order there so you have size on the book, you have to protect 16 the person who has provided the liquidity and if I want to 17 reach through and grab that liquidity by definition I should 18 say I forego wanting price improvement, I have made the 19 decision I want to buy that stock. I don't want somebody to 20 step ahead of that and free ride that person's order so. By 21 definition I don't want them to sell it to me cheaper and I 22 don't want them to go along with that purchase or I won't put 23 an order down there myself. 24 MR. BURKHART: I agree with Scott and to invert the 25 perspective, what we would, we, being the institutional asset 1 management community, would be used for when we put our large 2 and show large size orders on the limit order book is a free 3 option. It's a free option for the specialist in the New 4 York Stock Exchange, it's a free option for the floor 5 brokers, it's a free option for anybody that wants to step in 6 front of us for a penny or a smaller increment if we ever get 7 to that, to trade ineffectively and to dilute or shut us out 8 effectively. It's one reason why Scott is trading upstairs 9 and why we do and Minder does more today than perhaps in the 10 days of old, is we don't want to give that free option to all 11 the market participants. That's a structure issue. I'd be 12 more than happy to put 100,000 shares of ABC stock down on 13 the order book. I don't want to show it and I want to know 14 that I've got price/time priority with that order down there. 15 MR. DeSANO: I just want to add, the concept of 16 price improvement in a market environment in the past was 17 widespread I think made more sense today with decimals, I go 18 back to best execution is in the eye of the beholder. We 19 need liquidity, we need access, too. I think price 20 improvements should go to those who have orders out loud. If 21 somebody's on the book at 20.03, 20.04 and there's a block 22 out at 20.05, I want to reach out to 20.05, I'll pay the 23 person at 2, 3 and 4 cents at 20.05 and I want that liquidity 24 on the book. That's actual price improvement, not what we 25 have today. 1 MR. COLBY: Gunner, if I heard you correctly, you 2 said you'd be perfectly happy to put a 100,000 share order on 3 the book, but you don't want it to be shown, and you want 4 price/time priority. 5 MR. BURKHART: Does that sound contradictory? Is 6 that what you're asking me? 7 MR. COLBY: It's what everybody wants, to have your 8 cake and eat it, too. The question I'd ask is right now, we 9 have individual markets set their own price/time priority, 10 and forgive me if I'm encroaching on your panel here, Larry, 11 but there's no market price/time priority, and even within 12 markets the amount of price/time priorities vary. So my 13 question to you is how much payback would there be if there 14 was greater price/time priority within a market, and someone 15 mentioned the penny versus the larger increment, and also 16 there's a question that even within a market you might have 17 price/time priority but not across a market. Janet runs I 18 think a pure price/time priority system and that's been a 19 selling point of yours. 20 MS. ANGSTADT: Yes, it has been. But we certainly 21 have people who want to be able to access quotes with the 22 price/time priority. We have some order types that help 23 reach out to people who are very fast if we don't have 24 liquidity on the system, but it is a strict price/time 25 priority system within Archipelago. 1 MR. NIEDERAUER: I think what Scott and Gunner and 2 Minder are all saying, would be, I think the genie is out of 3 the bottle on this one a little bit. I think now you're in a 4 decimalized environment, we might aspire to have a pure 5 price/time environment, but under these conditions there's no 6 such thing as a pure price/time environment now. 7 When you listen to Scott's depiction of what a 8 larger institution would like to do, that's what I think the 9 NYSE Institutional Express order was designed to do, and we 10 could argue in practice whether it has or hasn't worked. I 11 think what you're seeing from our seat is competition is 12 kicking in. I think six months or a year ago we would say 13 let's mandate that since the NBBO is becoming less relevant, 14 but it's still important, let's just mandate that people have 15 to show more than one level. Well, it turns out you really 16 didn't have to mandate that, because everybody is showing 17 multiple levels. Folks like us are consolidating everything 18 that people are contributing and putting in front of 19 institutions like Minder, Scott and Gunner represent. On top 20 of that the competition kicks in with these various order 21 types. It may sound like these guys want to have their cake 22 and eat it too, I'm not sure in a decimalized environment 23 what else we're to do, other than to display a small quantity 24 display, have a lot on research and count on the competitive 25 forces again to take over where intelligent order routers 1 begin to understand there are certain pools of liquidity that 2 there's more than meets the eye behind there. I'm not sure 3 how we can solve that. 4 I think short of turning the clock back to a 5 smaller minimum price variation, I think these are natural 6 outgrowths of a competitive market structure that maybe 7 aren't as dysfunctional as we think they are. 8 MR. MURPHY: Could I weigh in a little bit, another 9 observation with Gunner and Scott in a decimal environment. 10 I still think price improvement is a valid option. In the 11 opening statement, Scott, when you said that shareholder 12 value for your fund holders is your primary concern, why 13 wouldn't you, if it was offered to you, want price 14 improvement? That liquidity is still there, the perception 15 is that someone jumped ahead of that published bid or 16 offerings whatever it is, for liquidity. One misnomer and I 17 know NYSE staff can back this up. The specialist as far as 18 pennying is extremely insignificant, almost nil, as far as 19 penny jumping. Really where the pennying comes in on the 20 Floor is agents that represent the institutions, who like 21 Minder said earlier, you know, the transparency is a double- 22 edged sword. A lot of people that have some size to do are 23 careful about what they divulge and advertise on the national 24 best bid or best offer, but if a large institution has a 25 further interest, and gets price improved on a block of stock 1 while that advertised, say up a penny or two, is still there, 2 if he has a further interest, it's still available to him. 3 So why not take advantage of that price improvement, even if 4 it is a penny or two? I don't think that's a bad thing. 5 MR. COLBY: Let's now shift directions and talk 6 about market data revenues. Currently, the public market 7 data system is very closely linked to self-regulation. SRO's 8 collect the quotes, the trade information from their members, 9 and they make this information publicly available, and 10 historically the SRO's have charged for this information. So 11 let me start with what was our Chairman's question at the 12 last hearing which is, is market data a public good that 13 should be available for free to investors and market 14 participants? Start with that one. 15 PROFESSOR KYLE: I'll answer that. 16 MR. DeSALVO: I don't have a problem with that. 17 PROFESSOR KYLE: There are a lot of externalities 18 that exist with respect to markets, and if you think about 19 market data, market data is providing a positive externality 20 to the people who use it. It has characteristics of a public 21 good in that once it's out of the bag it's a little hard to 22 prevent it from circulating around, but it can be prevented. 23 If you think of market data as providing an externality and 24 if you thinking about providing a limit order as hanging 25 yourself out there to get picked off, that's providing an 1 externality. That suggests that market data should be 2 expensive because the people who are benefiting from it as a 3 positive externality should in some sense maybe pay for it, 4 like a little tax on them having to pay for it I think is 5 probably reasonable. If you think about an efficient funding 6 for exchanges, market data fees, that is payments for a 7 positive externality that they're creating for everybody else 8 should be a large component of their revenues and that's kind 9 of an efficient way of creating that positive externality. 10 COMMISSIONER PITT: Now, let me follow up with 11 that, because there are different types of market data and 12 maybe the problem of coming up with an answer to the question 13 exists when you try to conflate the different types of data. 14 Let's just deal with transactions and stated bids 15 and offers. If you assume the best execution, whatever it 16 means, is a legal obligation and if you assume that the only 17 way you can fulfill that obligation is to be able to have 18 sufficient data to note where best execution will take place, 19 again, whatever best execution is, then if you just limit 20 market data to that kind of information, people who are 21 willing to have limit orders on the books in size, et cetera, 22 why isn't that akin to IBM's results of operations, for which 23 we do not charge; that's available to investors for free, 24 because it's thought to be relevant information about making 25 a judgment. Why isn't that equally true of market data about 1 bids and offers and actual transactions? 2 PROFESSOR KYLE: I think because in the case of IBM 3 providing set financial information, the shareholders of IBM 4 can be located as the individuals who ultimately will benefit 5 from having that information provided, and it's fair enough 6 to make them pay for it, so in some sense, maybe it's IBM 7 that bears the cost of creating its own financial information 8 and that's a reasonable way to finance the provision of that 9 information. But when you're talking about a trading system, 10 there a