UNITED STATES SECURITIES AND EXCHANGE COMMISSION ADVISORY COMMITTEE ON MARKET INFORMATION Thursday, April 12, 2001 9:07 a.m. Securities and Exchange Commission 450 Fifth Street, N.W. Washington, D.C. Diversified Reporting Services, Inc. (202) 296-9626 2 PARTICIPANTS: MR. MICHAEL ATKIN Vice President, Financial Information Services Division, Software and Information Industry Association MR. ROBERT G. BRITZ Group Executive Vice President, New York Stock Exchange MR. ANDREW M. BROOKS Vice President, Head of Equity Trading, T. Rowe Price MR. ROBERT COLBY Deputy Director, Division of Market Regulation, SEC MR. THOMAS DAVIN Nasdaq MR. MICHAEL T. DORSEY Senior Vice President, General Counsel and Secretary Knight Trading Group MR. JAMES DOUGHAN Susquehanna Partners MS. CARRIE E. DWYER General Counsel and Executive Vice President, Charles Schwab MS. ADENA FRIEDMAN Senior Vice President Nasdaq MR. THOMAS HALEY New York Stock Exchange MR. DAVID A. HUNT Partner McKinsey & Company MR. GEORGE K. JENNISON Senior Managing Director, Retail Equity Group, First Union Securities Diversified Reporting Services, Inc. (202) 296-9626 3 PARTICIPANTS (CONTINUED): PROF. SIMON JOHNSON Sloan School of Management, Massachusetts Institute of Technology MR. DONALD C. LANGEVOORT Professor, Georgetown University Law Center MR. BERNARD L. MADOFF Bernard L. Madoff Investment Securities MR. BRIAN McNELIS Reuters America MR. MARK A. MINISTER President and CEO, Bridge Training MS. ANNETTE L. NAZARETH Director, Division of Market Regulation, SEC MR. EDWARD NICOLL Chairman and CEO Datek Online Holdings MR. PAUL O'KELLY Chief Operating Officer, Chicago Stock Exchange MR. GERALD D. PUTNAM Chief Executive Officer, Archipelago MR. PETER QUICK President American Stock Exchange MR. ERIC D. ROITER Senior Vice President and General Counsel, Fidelity Management & Research Company MR. JOEL SELIGMAN, Chair/Moderator Dean, Washington University School of Law Diversified Reporting Services, Inc. (202) 296-9626 4 C O N T E N T S AGENDA ITEM: PAGE I. Progress Report on Subcommittee 6 II. Presentation by CTA/CQ Plan Administrator 29 III. Presentation by Nasdaq/UTP Plan Administrator 64 IV. Presentation on FISD Market Data Policy Database 10 V. Discussion 116 VI. Public Comments 151 VII. Summary/Next Steps 154 Diversified Reporting Services, Inc. (202) 296-9626 5 P R O C E E D I N G S DEAN SELIGMAN: Let me call our fourth meeting to order. We have what will be, I suspect, a pretty full agenda. We're going to begin with four reports. And just by way of a couple of preliminary points, I'm going to ask people to hold their questions on each of the respective reports until the initial presentation is done. I am concerned that when we're complying with our charter and essentially limited to six public meetings, that I want to use the time efficiently today and cover all or at least as much of the agenda as possible. I'm concerned that if we have too many questions in the process it will slow it down too much. Second, in three of the four reports there are slides. Each of the various presenters is going to ultimately reproduce the slides and we'll circulate them to each member on the committee. Let me start with Don, who is chairing the subcommittee on alternative methods of consolidation, and just get a progress report. Don, as you all know, besides leading another subcommittee meeting before May 14th, our focus on that date will be how a multiple consolidator system ultimately could work. Diversified Reporting Services, Inc. (202) 296-9626 6 PROF. LANGEVOORT: Right. Thanks, Joel, and given all the work we have to do today, let me try to be very brief. As Joel mentioned, the subcommittee, which contains 11 members of this full advisory committee, met once on March 26th and will be meeting again this Monday. Our objective is to prepare an agenda for the May 14th meeting that will reflect the questions we think the full advisory committee needs to discuss and, hopefully, some sense of consensus -- to the extent there is any -- among the subcommittee as to ideas, conclusions or whatever that has come out of our deliberations. The first meeting was a very productive one. We divided our discussion into two parts: one, the technological feasibility of moving to a multiple consolidator model, secondly, the economics of moving to a competing consolidator model. On the first of these two questions we received very valuable reports from Tom Davin and Tom Haley on the existing SIAC system, how it works, what the technological challenges are of bringing various data feeds into a single consolidated stream and what that would mean to the process of having multiple consolidators facing that same task. Bottom line -- and I won't go through all the details, we'll certainly present that to you in advance of Diversified Reporting Services, Inc. (202) 296-9626 7 the May 14th meeting -- it was a fairly optimistic one. That is to say, technologically, the process of receiving the feeds, sequencing, validating from multiple sources using multiple consolidators involves something of a challenge and will require some degree of SEC oversight. But the subcommittee is optimistic that the self- interests of their various participants, the experience that we've had in other areas -- vendors under the current system, options, commodities, other areas -- suggests that those technological problems would be worked out with a relatively small degree of SEC supervision. Thus our conclusion, our consensus, I think, is that the system is technologically doable. The economics poses a more difficult problem, and was the substance of our discussion at the second half of the first meeting and will continue to be the subject of the discussion on Monday. To the extent that a display rule is kept in place -- and we took it as the consensus of the full advisory committee at the March meeting, there's a strong feeling the display rule should stay in some form -- questions of market power, who can control pricing, whether there will be exploitative pricing, were serious ones. Obviously, it's a matter that divides members of the advisory committee and the full committee. The New York Stock Exchange has articulated Diversified Reporting Services, Inc. (202) 296-9626 8 numerous times that there are restraints built into the governance structure of the SROs and other mechanisms that prevent exploitative pricing. Not all members of the subcommittee or the full committee agree and one of the things we are exploring is whether there are compromises, middle ground options, that might be mechanisms for limiting that pricing power to the extent that it exists. What we've discussed -- and, to some extent, this anticipates the discussion this committee will have this afternoon -- are possibilities such as most favored nation pricing, which I think really means trying to figure out who is comparable to whom, so that they should pay roughly comparable prices. Secondly, the possibility of advisory committees informing the individual SROs with respect to pricing decisions, a supplement to the current governance structures that we have in place now. Perhaps -- and this is something we are going to discuss more on Monday -- maybe relaxing the display rule, perhaps looking at ways in which the time at which the NBBO or other data needs to be displayed is limited to time of execution rather than a broader requirement, or perhaps -- and I emphasize perhaps -- the possibility of allowing synthetic NBBO, NBBO that is comparable to, but perhaps not exactly the same as what it is now. Finally -- and, again, repeating a discussion that Diversified Reporting Services, Inc. (202) 296-9626 9 we've had previously on the full committee -- the possibility that various market participants can sell their data separately: broker/dealers, perhaps specialists, the exchanges, so that there will be multiple data sources competing in the marketplace. Again, we will continue all this discussion. So let me just finish up by saying, I think the subcommittee process has been very valuable. The smaller group and the informality of the structure has, I think, led to a good quality of communication among members of the subcommittee, and we're happy, I think, that we've made some progress, even if we haven't solved all the problems. So we'll see what we bring to the full committee on May 14th. I'd be happy if other members of the subcommittee want to add anything to what I said, but I think that's my progress report. DEAN SELIGMAN: Okay. It sounds very much like a work in progress and we'll have, obviously, a full exploration of the issues on the 14th, and, as you suggested, preceded by certain documentation. Let me go a little bit out of order for the presentations this morning we met to prepare, and I'd like to start with Michael Atkin and I'd like to walk through this FISD market data policy database and certain questions that suggests. Diversified Reporting Services, Inc. (202) 296-9626 10 With each of the presenters, I asked, in effect, that their presentations be fewer than 30 minutes. In the most optimistic world, the questions also would be within 30 minutes. If we need to discuss them further, we will. We'll see that some of what Michael will discuss will be related to then what is picked up by Tom Haley and Tom Davin for New York and Nasdaq. Michael, let me turn the floor over to you. MR. ATKIN: All right. Thanks, Joel. And I have a time line down to 15 minutes for you, so -- At the conclusion of the meeting, March 1, there was seemingly a lot of uncertainty and a lot of lack of clarity, if you will, about market data management, how the process worked, what was it. So after that meeting, the Commission staff asked me if I would put together a small paper, which I did, circulated it to everybody on this advisory committee, and I'm going to briefly summarize what was in that paper. My intent was to give you a neutral assessment, without an undue amount of judgement or assessment on my part, and to give everybody a common starting point so that you could have a conversation on market data policy issues even if you weren't directly involved in the wonderful world of market data administration. All right. Throughout the paper that I sent, I Diversified Reporting Services, Inc. (202) 296-9626 11 used the term "market data management" to refer to a whole spectrum of issues associated with market data policy, contracts, billing and reporting, and administration. However, I will make an important distinction between the issues that are process-related versus those that are policy- related, and I think that's a very important distinction. First off, the entire global industry is in support of and is actively working on process efficiency. The goal is to promote electronic commerce to reduce the costs associated with market data administration. This is clearly a classic win/win for everyone. Still, there are lots of inefficiencies in the system that can be improved and the industry recognizes them and is cooperating to address them. Business policies of the exchanges are a different matter. There is no clear consensus on the right business model. There is no clear right answer. There are honest differences of opinion on what it takes to run the business of the exchange and there are many shades of gray. If you'll bear with me, I'm going to try to shed a little bit of light on why things are so gray. I think it's also important to recognize this is not a U.S. problem. There are over 85 fee liable exchanges around the world. There are dozens of third party services that all work together to make up the market data inventories of the vendors and the user firms. There are lots of Diversified Reporting Services, Inc. (202) 296-9626 12 policies, lots of variations of policies, and they all need to be understood and applied by organizations that are in the market data redistribution business. Unfortunately, administration is part of the job, and a confusing job it is, because not only do you have to comply with the multiple policies of hundreds of sources, but you've got to apply old contractual requirements to real world circumstances. You've got to do so for situations that were neither known nor anticipated when the policies and contracts were created, and if you really think about it, that's probably the best reason why all of you should go back to your organizations and dramatically increase the salaries of your market data managers. This being April and I'm Jewish, I had to do the Amonish Tonok slide. Underlying the entire discussion, I believe, are four core questions. The first is, who should determine the business practices and business models of the exchanges, particularly in this rapidly evolving industry environment. The second, I think picks up on Don's comment, is to recognize that the generation next, meaning the critical issues under debate within this committee, transparency and display rules and the like, are very different from the business issues and business structures that currently exist. The question I think that everyone is asking is, how do you Diversified Reporting Services, Inc. (202) 296-9626 13 maintain a level playing field and ensure that the business practices of the next generation of services don't have the unintentional consequence of impeding commerce. The third is what, if anything, outside of what the industry is doing cooperatively, can be done to streamline, simplify, standardize and automate business processes, in essence, to promote the straight through processing of market data administration. It's a shared goal amongst everyone and it's one that has real dollars associated with it and potentially will save the industry a lot of money. Then ultimately, and I think the reason we're -- or partially the reason why we're sitting in this room -- is what's the practical role for the Commission to help the industry manage those business challenges? To state the obvious, all market data management questions start with the contract. The market data contract is the principal document that governs what vendors, redistributors and subscribers can and can't do with market data, because once it's signed, all parties are obligated to comply with the terms and conditions. Not surprisingly, our members spend a lot of time dealing with contract issues. In fact, we develop, we publish, we continue to evolve something we call our Exchange Contract Guide to help support this delightful process of contract negotiation. Diversified Reporting Services, Inc. (202) 296-9626 14 I don't want to be too flippant about it. I think it's a very important document. It's freely available on our Web site to the entire industry. Its purpose is to dig deep into the bowels of the contract issues by an unbiased analysis of what the business considerations are for each of the segments of the contract, and then we illustrate that with real examples from market data contracts around the world. Every contract, in one way or the other, deals with these 12 core market data issues. What we did was to combine all of the model agreements from all of the vendors and all of the contracts that existed in the industry into one big document, get everybody discussing this document to define the points that are in it, and to make sure that we could do two things. Not only do you understand the objectives of each of these contract provisions, but you've got to look at the practical problems associated with implementation. The result is a sample contract. It's got lots of analysis. It's all used as a starting point for negotiation. This document is being used and has been used for well over a year in the industry. In fact, we're now getting pushed to add more to it, add redistribution issues associated with new things people are doing, index redistribution and stuff like that, re-derive data. So when you think about it, when you think about Diversified Reporting Services, Inc. (202) 296-9626 15 our policy database, it's all driven off of this contract structure. So what's the problem? Well, besides the point that there are some -- forgive me -- very draconian contracts out there. The rules related to market data are governed by this legal document. This is a legal document that's difficult to negotiate. It's also subject to interpretation. Understanding and translating the complex policies of multiple exchanges is anything but simple. Because they're legal documents, they don't necessarily spell out how the rules are to be implemented in the real world. Because they are complex and they're difficult to negotiate, they have a long life span, for years, many of them. As a result, they're out of date, they don't address the new technologies, they don't address special circumstances, they don't deal with new applications. Contracts are often vague, subject to interpretation, which, of course, leads to conflicts of interpretation. Compliance with the rules has also got to be communicated. You've got to deal with the various people who work with data all throughout the information chain. These rules are complex. They sometimes conflict with the internal objectives, the sales objectives of the people who are redistributing data, and it's the lack of knowledge and that breakdown of communication that I believe leads to the Diversified Reporting Services, Inc. (202) 296-9626 16 downstream errors, which sometimes result in significant financial liabilities. Then finally, as mentioned, the contracts vary around the world. Keeping track with all the obligations, particularly as it relates to redistribution, and especially with nontraditional vendors in the business, is a difficult and an error-prone process, which is why we built our market data policy database. Our objective is to document all of the obligations and requirements covered by exchange contracts and other policy documentation at a very granular level. In fact, we have taken our 12 contract segments and have now translated that into about 230 individual contract type of questions that you can ask. The database now includes agreed market data policy statements from 45 exchanges. We are going to completely cover the industry, and once we finish the exchanges, we're intending to do the third parties and vendors, and we've done fee schedules that have been normalized for 75 of those exchanges. The purpose of the database is fourfold. First, know thy rule and comply with the requirements. Ignorance of the rules should not be anybody's excuse for not following the rules. You might not like them, but you've got to understand them. Second, make sure that the vendors and their Diversified Reporting Services, Inc. (202) 296-9626 17 downstream clients have consistent, complete and uniform interpretations of all the administration requirements on a global basis. In essence, anything that's ambiguous from an exchange needs to be clarified because you're holding people accountable for following the procedures. Third, make sure the polices are rational. Lots of the contracts are outdated. New policies need to be created. In essence, we built this to allow the industry access to the collective experience of the global market data industry as either they had to update their contracts or develop new ones. So wondering about Internet redistribution policy? Go check the database and see how others are doing it. Trying to figure out how to deal with creating an index using proprietary exchange data? Go check the database and see how others are doing it. And hopefully, rationality wins out and you tend to move toward uniformity. In essence, any debate on policy starts with what's the current policy? Then, how does that policy compare to others so that you can have a rational conversation amongst the various parties on what makes sense in the long run? Finally, as more and more people get into the market data redistribution business, they need some place to reference their requirements. What are the requirements and costs to redistribute real time data from the New York Stock Diversified Reporting Services, Inc. (202) 296-9626 18 Exchange, EuroNext and Deutsche Borse via my Web site? What kind of agreements are needed if I want to take a snapshot of data, redistribute it via my Web site, using these eight European exchanges and Chicago Board of Trade? That kind of thing. We are now in beta tests. We are running the thing through its paces. Our purpose is to make sure that real people who work with market data can ask real plain language questions and get useful answers out of our database. All right. That's enough on the tools that we're trying to build. Let me turn to the business side of the equation, because this is really, I think, getting into the core of the question on who should determine the business practices of the exchange. The first, this prior approval versus vendor discretion, is really a debate about the business models that exist in the industry. These are the two primary models that currently exist. Vendor discretion, which is used by most of the exchanges around the world, gives the vendor a license to redistribute data without prior approval, but subject to the terms of the exchange contract. The prior approval model starts with the premise that no one is allowed to do anything with the data until it is approved in advance by the exchange. Vendors say the prior approval is administratively cumbersome. The North Diversified Reporting Services, Inc. (202) 296-9626 19 American equity exchanges say, this is necessary for us to effectively manage our business. I'm going to get into that in a second, but it's not clear, from my experience, that there is an easy answer to that dilemma, or actually to any of these dilemmas I'm going to paint for you. North American equity exchanges use various classification systems to determine the rights to use data and the fees associated with usage. For example, there's a distinction between professional and nonprofessional subscribers for fee determination. There are distinctions between vendors, subvendors and subscribers for assigning contractual and liability obligations. There are distinctions between internal redistributors, external redistributors and no redistribution for fee determination and for assignment of administrative requirements. Unit account: it's critical because this is, in essence, what you're required to track and what you're required to report on. Units can be device-based, user- based, location-based or I.D.-based. There are unit account policies for single users, for common areas, for shared terminals, for multiple terminals at a single desk. This is an area of huge debate. Do you count units, passwords or locations? If there are two data feeds supporting one user, is that one unit or two? What about if that application is serving both Diversified Reporting Services, Inc. (202) 296-9626 20 trading and analytics? Is that one unit or two? Each exchange sets its policy for how it both determines its unit and applies those policies. How about for enterprise licenses? What's the basis? What's the metric to use? Is it accesses? Terminals? Registered reps? Trading turnover? What's the metric? These are just not simple questions and there's more than one side of each of these issues and I wanted to make sure that as we were talking about them, they were characterized correctly. Billing and reporting is another huge area of debate. Billing and reporting right now is -- I would characterize it as a multiple system, costly, manually intensive, administrative nightmare. The good news, however, is that everybody agrees with that and everybody is on board with the objective of simplification and automation. But it's a huge task. It's complicated by differences in both reporting requirements and differences in billing approaches. Let me give you an example. For data feeds, the New York Stock Exchange, Nasdaq and OPRA maintain a direct contractual relationship with their subscribers. They have this prior approval process through the Exhibit A for New York and OPRA and Attachment A for Nasdaq. The vendors don't like it. They think it's cumbersome and unnecessary. They want the exchanges to adopt Diversified Reporting Services, Inc. (202) 296-9626 21 a model that says, authorize the use of data subject to your terms and conditions. The exchanges, on the other hand, have adopted this convention because they need to have it in place to support that direct contractual relationship with the subscriber, which begs the question, why are the exchanges in the billing business? And, in fact, a lot of vendors ask that. Why are the exchanges in the billing business? Vendors say, "Get out of the billing business." But not everybody agrees with that. The exchanges and the large user firms like direct reporting because it makes reconciliation easier and it makes billing more accurate. The redistributors are not necessarily the best in doing these administrative processes on behalf of the exchanges. Some vendors have said, "We don't want the burden and the cost and particularly the liabilities of being the billing agent for the exchange." Again, not a simple issue to resolve. And then there's also -- I don't mean to make them any less significant -- issues associated with intellectual property rights for delayed data and derived data. What's the extent to which vendors and subscribers can create their own intellectual property using exchange market data? I think this is important as we begin to look at the new world of exchange competition, or to the extent to which this data is subject to fees and reporting obligations. Diversified Reporting Services, Inc. (202) 296-9626 22 So what I wanted to do was just paint for you -- and what I tried to do in my paper -- what I thought were the seven core business issues that exist. The bottom line point of all of that is it's not a simple subject to figure out who sets those policies and how those policies are set. Let me talk a little bit about the good stuff. There is a lot of coordination and cooperation around the world on business process automation. While it's a massive undertaking, I think the payoff is potentially very significant. And I frankly want to give the exchanges their due on this, because they are not only active and willing participants in this discussion, they are often the lead in moving this activity forward. The system, you know, the evil, hideous Exhibit A process, is a great example. It started out with multiple agreements. Nasdaq had their own. New York had their own. OPRA had their own. Toronto had their own. Everybody said, "This is crazy. We've got all the same stuff." The exchanges agreed. They got together through our organization. We created a single document. That document is currently in use. But that's not all. The exchanges are now taking it a step further. Two weeks ago I spent a couple hours with the New York Stock Exchange, took a look at what they call Diversified Reporting Services, Inc. (202) 296-9626 23 their Automated Data Feed Reporting Process. It's a beautiful system. It is a Web-enabled Exhibit A, reduce that whole process to a series of click-throughs, automatic validation, checking reference available to the vendors. It's really sweet. I think it will significantly reduce the turnaround time for authorization and will allow faster response to marketplace needs. Now it begs the question, of course -- still open question -- is that process needed? But that aside, the process that exists is moving toward electronic commerce. Also on the docket -- I talked to Ron Jordan over at the exchange -- is click-through subscriber agreements for professional subscribers. This is something that Nasdaq led the way in. They now have it implemented. Nasdaq started it. The New York Stock Exchange is following suit. I think that's a good example of how this process works. The New York Stock Exchange was the first exchange to adopt the user definition of unit account. Users all got together, tried to come up with a standard contractual definition of unit account. NYSE said, "It makes sense to us. We will implement it." They eliminated double billing and are no longer -- if you get New York Stock Exchange delivered to you by two vendors to a single desk, you only pay once for that. That's their policy. And they've established procedures to make sure that firms get credit and Diversified Reporting Services, Inc. (202) 296-9626 24 can take advantage of that. The exchanges are also helping achieve and willing participants in the automation of reporting, the simplification of billing processes. I think this is a hugely important initiative, and will significantly reduce costs. So as you look at this equation, understand what's going on right now in the industry. There are legitimate business issues under debate and everybody is working together, in my opinion, to try to make the process more efficient. All right. Let me sum up. First point is, do not, please, underestimate the complexity of market data management or try to simplify these issues as we slog through this debate. I'd even go so far as to say, the problems that are being articulated by the industry are not about fee setting and they're not about SEC regulatory oversight. They are more about these market data challenges that I've tried to lay out that the industry is debating right now. Second, everyone seems to be in support of market data contract and fee policy transparency. Every query that we've looked at says, in order to fix these problems or to have a debate on these problems, you've got to find out what the current policy is, and everybody is willing to play this game. I don't think transparency is a question. Third, the entire industry recognizes the Diversified Reporting Services, Inc. (202) 296-9626 25 complexities associated with market data management and they are working cooperatively to try to simplify, rationalize and automate the market data business processes. Business process automation is a huge win. There are lots of opportunities for simplification, consistency and uniformity. I think we might be well served to try to figure out how we can support activity, because it has got a lot of inertia and it's a difficult thing for the industry to get its arms around. Last slide: I have been listening to the discussion on business models and on market data policy for a long time. The right orientation on these issues is still not completely evident. All sides make legitimate points. There is a productive dialog taking place on business policies. We're working to try to build some tools to allow the industry to have some empirical data so that they can make their case. At the end of the day, however, it seems to me that these are legitimate business issues that need to be managed on the basis of the business realities of the involved parties. The direct involvement of the Commission is likely only required if one group is being obstructionist or if there is an impediment to commerce. I personally don't think that's happening, but I'll leave that judgement to others on this advisory committee. Diversified Reporting Services, Inc. (202) 296-9626 26 As I indicated in my paper, the discussion concerning the level of information required for market transparency, including the concepts of information competition, the role of exchanges as value-added information providers, the notion of mandates on the factors of production, and the potential for changes to the display rule, I believe, in my conversations with the industry, are the most important points that we are debating and the ones that we would probably be well served to devote our time on having a full debate. So with that, Joel, I'll take any questions if there are some. DEAN SELIGMAN: Thank you, Michael. And let me just -- to be sure everybody appreciates what, in part, your database will achieve is increased transparency of contractual terms, including on a global basis, increased transparency of fee schedules. Individual exchange user contracts, including Exhibit A, would not be captured in the database, and who is subject to particular aspects of the fee schedules is not captured in the database. MR. ATKIN: Yeah, I would think that that's proprietary information that the firms don't want to have shared with their competitors. I talked to one firm, for example, that's doing a new transaction-based system and is trying to figure out what Diversified Reporting Services, Inc. (202) 296-9626 27 its market data policies are. They sent me a note and said, "Please don't share this with anybody because we're trying to get a jump on the market." So I don't think that's really an exchange issue, I think that's really a vendor and end-user issue. DEAN SELIGMAN: When do you think you'll be done with beta testing of the database? MR. ATKIN: It's hard to say, because the purpose of the beta testing is to make sure it's right and we're going to do it until it's right. DEAN SELIGMAN: Sure. MR. ATKIN: It will probably be finished by June or July of this year, enough that -- I mean right now I'm pretty confident that the data is accurate. What I'm a little bit unclear about is our retrieval interface in trying to -- because this is complex stuff and you've got to figure out how to do multiple queries on various policies. So we're working on our interface to make sure that the end user ultimately who's making the decision about, do we do something, can get the information they need. So my guess is mid-year. DEAN SELIGMAN: Okay. Are there other questions from the committee? Okay. Michael -- oh, I'm sorry. Bob? MR. COLBY: Can I just ask, within the U.S. Diversified Reporting Services, Inc. (202) 296-9626 28 context, do you have a view as to whether having multiple consolidators of the data as opposed to the consortium, how that would affect the data management issues? MR. ATKIN: I don't think it would affect it at all. I think the rules and policies are set by the data originators. The distributors have to follow the rules, but having multiple distributor points I don't think is going to change those rules at all. MR. COLBY: Right now I think the policies are set by the administrators on behalf of everyone in the plan and if they were being set individually by each of the members that currently participate in the plan as opposed to by one single consolidator, would that affect that? Is that -- MR. ATKIN: I might be wrong here, Bob, but I'm not sure if the first statement is 100 percent accurate. I think the policies are set by the exchanges. The plan administrators just administer the policies. I mean, Tom, you probably know a lot better -- MR. HALEY: That is correct. MR. ATKIN: Yes. DEAN SELIGMAN: And we will certainly amplify that point in presentations that are coming. Other questions for the group? All right. Well, Michael, thank you very much, and I have a feeling your thoughts on this will be interwoven Diversified Reporting Services, Inc. (202) 296-9626 29 throughout today and the sessions to come. Let me turn to Tom Haley now to focus on the CTA/CQ plan administration. There are a number of specific questions that were propounded, and I think you also have a PowerPoint presentation? MR. HALEY: I do. DEAN SELIGMAN: Okay. MR. HALEY: Thank you, and I believe that the way we're going to do this, this morning, is that I'm going to do the first half of the 30 minutes and Bob is going to do the second half of the 30 minutes. I've got some slides that will set the stage as to what we do as the network administrator and Bob is going to cover a number of the questions that have been raised. DEAN SELIGMAN: Fine. Why don't you proceed? MR. HALEY: Okay. Well, let me just start right out here with the things we're going to cover. I flipped through that slide and we don't have a backup on here, but we're going to cover -- basically this first slide, I just wanted to set some perspective. What we're talking about here for everybody, just to jog your memory for a second, we've got two plans. We've got a CTA plan and the CQ plan. CTA covers transactions. CQ covers quotations and the BBO. The SROs that are participants in those plans govern those plans. There are currently nine SROs that Diversified Reporting Services, Inc. (202) 296-9626 30 participate in the CTA plan and the CQ plan. Under those plans there are two networks: Network A and Network B. Network A refers to trades and quotations in respective NYSE listed equities. Network B refers to trades and quotations in respective AMEX and other regional-listed equity securities. We're going to be talking about the Network A administrator today. There is also the -- AMEX administers Network B. And I think, as you well know, there is one processor, SIAC, and there are two systems that SIAC operators, which are the consolidated tape and the consolidated quotation system. And I mention those two because that's really the starting point where the contract process begins, when we get back in here in this presentation to the contracts. NYSE is the Network A administrator, as you know. What does NYSE do? Well, basically, it carries out the duties and responsibilities pursuant to authority that's been delegated to it by the SRO participants. So we operate pursuant to delegated authority. That authority was delegated to us by the participants. NYSE, in terms of carrying out its Network A administrator responsibilities, is obligated to administer Network A agreements and rates as established by the participating SRO boards and approved by the SEC. So all of Diversified Reporting Services, Inc. (202) 296-9626 31 our contracts and rates are part of the plans. They've been approved by the SROs, filed with the Commission and approved by the Commission. What do we do on a day-to-day basis, to get into more in the way of specifics as to what our role entails? Basically, it's contract administration. We contract with vendors and other data recipients for access to Network A and Network B data. The data that comes out of the consolidated tape and consolidated quotation system comes out in a single stream containing both Network A and Network B data. We do the contracts with folks that want to interface with the consolidated tape and the consolidated quotation system and we do that on behalf of Network B as well. We also contract using standard form agreements with subscribers for Network A data and we create and oversee Network A pilot programs. The administrator bills and collects appropriate fees as per the Network A rate structure. One aspect of our duties used to be to arrange for leased lines to support the Network A ticker. That's pretty much phased out at this point. That's the low speed visual ticker. We also oversee SIAC's operation of the consolidated tape and consolidated quotation system on a daily basis. There's coordination between NYSE personnel and the SIAC personnel to the extent that it's necessary and appropriate, Diversified Reporting Services, Inc. (202) 296-9626 32 on a monthly basis. We pay SIAC's bills for their production and development. We also distribute the net revenues to the participants and provide an accounting to them. The next chart is a single page. This rate structure is the rate structure that is in place for display fees, that has been approved by the participant SROs, and which has been submitted and put through the regulatory review process, approved by the Commission. This rate structure is also readily available. This rate sheet is readily available on the NYSE'S NYSEData.com website, and the rates that you see up there generally generate and account for 90-percent of the network A revenues. Most of that comes from the top part of that sheet, which is the professional display fees for professional subscribers. As you can see, it's a tiered system. This was -- it's got 14 tiers. It was established in 1987. During the interim between 1987 and now, this -- there's been one 6-percent increase that took effective January 1, 1992. We maintain a database that has over 15,000 professional subscribers in it, professional subscriber accounts, 500,000 display devices on a global basis. And at the end of the day, on a monthly basis, this boils down to roughly an average unit cost of about $25 per month per terminal. These fees also entitle a recipient to Diversified Reporting Services, Inc. (202) 296-9626 33 receive and display last-sale prices, transaction reports, bid and ask report, bid and ask prices, ticker, all on the same screen, together with BBO and Montage. We do not have separate pricing for BBO and Montage, it's all-inclusive in the rates that are paid for network A display. The second part of that rate structure that you saw up there previously was display fees for nonprofessionals. The nonprofessional rates are, basically we have a monthly fee, and for data that's made available to nonprofessionals, these fees will apply to the vendor providing the data, whether that is a traditional form of vendor, such as a Reuters, or whether or not it is a broker-dealer such as a Schwab, or a Fidelity, or Soloman Smith Barney, or what have you. The fee for the first 250,000 subscribers is a dollar, and in excess -- over 250,000, it's 50 cents each. If you're serving your subscriber base, or your customer, your retail customer population on a pay-for-use basis, or on a usage base, rather than pay a monthly fee, we have established usage rates. Those usage rates are what you see there. The vendors meter the data. And for the first 20 million quotes, the fee is three quarters of a penny. The next 20 million, it drops to a half a penny, and for over 40 million, it drops down to a quarter of a cent. And these two are tied in. If a vendor Diversified Reporting Services, Inc. (202) 296-9626 34 that is using usage-based has a nonprofessional subscriber who is a heavy user, and he is taking down streaming data, that individual is capped for purposes of what the vendor pays, at the monthly dollar rate. You've heard a lot about nonprofessional fees over the course of your meetings here, but I just put up this slide to share with you some of the histories. That displays the monopoly power of the New York Stock Exchange and the Network A consortium. In 1984, a nonprofessional paid a hundred and sixty eight bucks for a display of Network A data. Today it's down to a dollar. Lastly, we recently put in place what we call an, "enterprise maximum," and that is for broker-dealers that are providing services to their customers, or broker-dealers with in-house devices. The maximum that they will pay on a monthly basis is $500,000 in respect of their in-house devices, or in respect of what they make available to their retail account holders. The two combined, their liability will not exceed $525,000 for the month. That was put in place in 1999. That, in effect, says that a firm, for instance, Carrie's -- Carrie knows the -- Schwab's got seven and a half million accounts, from what I read. All of those accounts could be serviced with streaming data on a monthly basis, and Schwab's bill would virtually not go up any significant Diversified Reporting Services, Inc. (202) 296-9626 35 amount, and certainly not go above five and a quarter. Let me turn now to agreements. The Network A agreements that the administrator has to administer are also agreements that have been put through the regulatory review process, that have been filed with the Commission, and that in turn have been approved for use by the Network A administrator. And again, to reiterate, the Network A administrator -- these are the only agreements that we have, these were approved by the SRO boards, by the Commission, and these are the agreements that we use. They're readily available to vendors, investors, via the NYSE website. They apply uniformly to all data users. And basically, we have two standard forms of agreement. One for -- which we call the, "vendor form" of agreement, and another called the, "subscriber form" of agreement. The vendor form is the standard form of agreement that all organizations that interface with the SIAC systems, or that interface with a vendor to get a data feed service must enter into with us. Now, Mike spent a second talking about preapproval. Part of that preapproval process is that if Reuters gives a data feed to one of its accounts without our preapproval, Reuters remains liable for the actions of that account and their use of the data. If they get our preapproval, they're off the hook. So therein lies part of Diversified Reporting Services, Inc. (202) 296-9626 36 the logic behind preapproval. The vendor form of agreement also incorporates exhibit A, which I'll talk a little bit more about in a second. The subscriber form of agreement. Basically, we only have, for professional subscribers, a standard written form of agreement. I was surprised, Mike, but I'll take your word for it. I didn't think we let Nasdaq ahead of us with a click-on for professionals, but apparently we have. In the nonprofessional arena, we have a standard form of written agreement for nonprofessionals, but over the years, we've also adopted, in addition to that -- substantively the same, but somewhat different in format, we allow vendors to take and add an addendum to their contract, rather than have to use ours today. But basically, the terms and conditions in that agreement are the same. And we also have developed a click-on agreement that the vendors can use with nonprofessional subscribers. Briefly now, on the exhibit A, as I said, the Exhibit A is an attachment to the vendor form of agreement. In the commercial context, I say this is somewhat analogous to a bill of lading or a packing list. This, the Exhibit A, is the understanding that we, as the Network A administrator, have with the customer, as to -- and it documents what the data is that they're getting, how they're going to be using it, what appropriate fees will be levied, what fees from the Diversified Reporting Services, Inc. (202) 296-9626 37 approved price list will apply, in this case. And I think, as was pointed out by Mike, and also I will add to it ,that we started out with Exhibit A, but subsequently AMEX, Nasdaq, OPRA, Board of Trade in Toronto, have adopted a similar concept. And that's it. That gives you the baseline for what the administrator does. DEAN SELIGMAN: All right. Let me -- before we address those separate issues, are there questions anyone would like to pose to Tom with respect to the role of the administrator? Yes, Mark? MR. MINISTER: Yeah, Tom, are there any current differences in data content or display restrictions between professional and nonprofessional? MR. HALEY: None. MR. MINISTER: There had been. I'm old enough to remember there used to be, but there are none now. MR. HALEY: I'm -- MR. MINISTER: You're not that old, I know. MR. HALEY: I -- well, I'm having a senior moment, I guess, because I don't remember. MR. MINISTER: So there are. So I guess the question -- and I don't know if you're the one to answer it -- why the average professional user pays $25, and the average nonprofessional user pays $1. So I'm trying to find the 25 to 1 difference, if there is no difference in content Diversified Reporting Services, Inc. (202) 296-9626 38 or display. MR. HALEY: Okay. I think if you go back to 1984, all there was, was one. There was no distinction made between pro and nonpro. One of the things that we've tried to do over the last 15 years is, promote real time for the nonprofessional, or the retail account holder. In turn, it's been, from our perspective -- and I believe the Commission has gone along with us -- reasonable and appropriate, and fair to offer this form of discrimination in our rate structure, in order to make more widespread real-time data to the individual segment of the investor community. DEAN SELIGMAN: I guess if I can just follow up Mark's question, are you running, in effect, a cross-subsidy? In effect, are the professional fees subsidizing the nonprofessional? MR. HALEY: Well, the professional fees continue to represent three-quarters of the total that is generated, in terms of Network A fees. I don't think the use of, "subsidy" would be the right word. We've got a pool of revenue that's being derived from different sources and uses of the data that we think are consistent with the objective of trying to promote more widespread dissemination of real-time to the benefit of investors. And, you know, you've still got the delayed issue sitting out there. And that's been something we've tried to overcome. Diversified Reporting Services, Inc. (202) 296-9626 39 MR. BRITZ: Joel, if I may, I think there are two answers. One answer is that, via consensus from the industry and, frankly, at the encouragement of the Securities and Exchange Commission, the notion of making a data product -- a real-time data product more widely available to, "individual investors" through favorable pricing, is something that the industry absolutely got behind and supported. So there's the whole consensus issue. But there's also a sort of more complex analysis of this, which you can make yourself crazy doing, but if you care to go down that path, a broker at average rate $25, sitting at a screen, can service a number of customers -- I haven't a clue what that number is -- via the telephone and via constantly calling up symbols. So what you need to do somehow, if you really want to get at to whether or not there are cross-subsidies here is, look at that number of customers that can be serviced by a single $25-dollar screen by a single, or perhaps multiple brokers, relative to the so-called nonprofessional click-on. MS. DWYER: Joel, could I just say, I think you're assuming that there is an issue of subsidy here, and yet -- DEAN SELIGMAN: I'm just posing a question. MS. DWYER: No, I'm saying, but I think that's the wrong question, because I don't think we've established basic cost of furnishing this. I think you should look at it as a Diversified Reporting Services, Inc. (202) 296-9626 40 pricing differential. And I would just say, for an average brokerage firm, it's not $20 a month, it's $20 to the New York Stock Exchange, $27.25 to the AMEX, $20 to the Nasdaq $50 if they get level II -- and $10 per terminal to OPRA, plus vendor fees on top of that. That's quite a pricing difference. DEAN SELIGMAN: Okay. Let me just pursue one other question. You alluded to pilot programs. How many pilot programs do you currently administer? And secondly, are there any pending? MR. HALEY: Bob was going to cover that, but I'll do it. We currently have one pilot program underway, and you've all seen it. It is the ticker display on CNBC and on Bloomberg. There are five TV networks that are participating in that pilot, and it's been going since 1996, and it's our intention to file that in the year 2001 with the Commission for permanent approval. There are none on the drawing board. DEAN SELIGMAN: Other questions from the committee? MR. JENNISON: I have one. DEAN SELIGMAN: Sure. MR. JENNISON: Tom, is that to say that there are no unique pricing scenarios out there with specific firms, other than the one you referred to? Other than that one pilot program, there are none out there? MR. HALEY: There is one Network A pilot program, Diversified Reporting Services, Inc. (202) 296-9626 41 and it's been that way for the last four or five years, four years at least. MR. COLBY: Tom, just for absolute clarity, do you have any ability to negotiate individual deals outside of the pilot program? MR. HALEY: Do not. DEAN SELIGMAN: So your fee structure is uniformly administered. MR. HALEY: We administer the fee structure that was presented there, and use the contracts. And some of you in the room perhaps have, from time to time, attempted to have your legal counsel make changes to those agreements, and we won't do it. They are the same agreement across the board for everybody, and NYSEdata.com has it all up there if anybody would like to see it. DEAN SELIGMAN: Eric? MR. ROITER: Thank you. I have sort of a reverse engineering question. The slide that you showed, that showed the dramatic decrease in pricing over the last 10 or 15 years or more, can you explain what factors accounted for the higher fees in the early years? MR. HALEY: Okay. Where we started out in the first year, in 1984, is that there was no distinction between professional and nonprofessional. We didn't make a distinction between subscribers, and we didn't have a Diversified Reporting Services, Inc. (202) 296-9626 42 separate category of subscribers. So everybody paid the same rate. That rate at that point would have been a hundred and sixty eight bucks. I think it was in the 1984 time frame, there was an outfit that was -- QuoTrek -- they were developing a hand-held device, and they came to us and asked for our help in setting up separate rates for nonprofessionals. They thought this was a technology that would take off, and bring real-time data to the individual investor. That resulted in the $13-dollar fee that you saw there, I think in 1984. In 1987, in conjunction with your organization and Carrie's organization, to combine last-sale bid/ask and ticker, which were nonprofessional -- which were separate rates for nonprofessional, into a single rate, we brought it down to $4. And there was kind of an understanding among OPRA, NYSE, AMEX, and Nasdaq, to see if we couldn't all somehow or another get nonprofessional into the area of the price of an eighth on a hundred shares, which would have been $12.50 a month. I think the actual thing came out to about $13.50. NYSE's piece of that was four bucks. It went up to 6 percent Jan. 1, '92. We had a modest increase in '97. And then, as a result of much dialogue with the broker/dealer community, in conjunction with the fact that the internet had come into play, there are a lot of customers out there now with PCs that wanted to get Diversified Reporting Services, Inc. (202) 296-9626 43 access to real-time data, we took, I think, some very aggressive action and reduced those rates substantially. And that's kind of the history of it. It was something that was a lot of dialogue back and forth between ourselves and the broker/dealer community, and to a certain extent, the vendors. If that answers your question, Eric. DEAN SELIGMAN: Other questions? Well, Tom, thank you very much. Let's turn things over to Bob, then. MR. BRITZ: I also apologize for the slides. I've got, really, just a bit of housekeeping to do up here. Annette asked us to address a handful of questions in our presentation that were originally contained in a March 20 memorandum from Carrie Dwyer, and while Tom, I think, has done a terrific job implicitly addressing some of those -- and I won't go over the territory he has covered -- Carrie has raised a number of very pointed questions, that I think are deserving of explicit answers, rather than implicit answers. So the words you see up there are Carrie's, the highlighting is mine. I'm just trying draw your attention, and I don't know how readable that is. I'm just trying to draw your attention to the -- trying to correlate the answers to the questions, if you will. The first question is, how exactly is market data information priced? What are the full extent of the cost Diversified Reporting Services, Inc. (202) 296-9626 44 burdens imposed on broker/dealers, market data vendors, taking into account exhibits and pilot programs? I won't spend a lot of time on this, because we have spoken and written extensively in terms of how market data is priced -- Tom has covered it, discussions with Mark and participants initially, to the extent that there is any traction, industry associations, consensus building, gauging the impact on SRO funding, balancing that against the objective of widespread data distribution. Vetting with the CTA, pre discussions with the SEC, approval by the boards, public filing, public notice, public comment. And the SEC does with it what it will at the end of the day. So that's essentially a snapshot of the first part. That dialogue has in recent years been a fairly easy dialogue, because pricing actions have all been of the south variety. It makes the conversations a little easier. That process I just very quickly described to you, it's exactly what happened in 1999. And maybe this gets to Eric's point as well. There have been two dramatic revenue cuts in the nonprofessional space;, 50 percent in 1997, and another 50 percent in 1999. I won't bore you with the 1997 ancient history. In 1999, we went through that process that I just described that led to that 50-percent reduction in the nonprofessional rates that Tom showed you earlier, and also Diversified Reporting Services, Inc. (202) 296-9626 45 created the enterprise arrangement, which has worked to the benefit of a number of broker/dealers. Picking up quickly on some of the highlights, there's a notion here that there are additional terms and conditions imposed, I guess, by the Exhibit A. I would suggest to you that Exhibit A is a printed form. It is a fill-in-the-blanks questionnaire form, the sole purpose of which is to elicit from the applicant what data they wish, how they'll use it, to what extent -- are there 15 display units? Are there 15,000 display units? -- from what source? Do they wish to drop a line directly into SIAC? And questions along those lines. They do not impose any additional terms and conditions. The irony here is that the notion that somehow what is contained in the Exhibit A injects variability into the application of our pricing, is truly an irony, because the sole purpose for the Exhibit A is to understand how the firm is using the data, and to what extent, and what data, and apply that to the published standard, uniform rate schedule, so that we know what fee A is appropriate for that particular firm, and to the extent that another firm comes in behind that firm, so that we can in fact produce a process that creates uniformity. Costs and burdens that vary from participant to participant. The answer to that is, sure they do, because if Diversified Reporting Services, Inc. (202) 296-9626 46 you want 15 units as opposed to 15,000 units, they're absolutely going to vary. But I think the term that is used repeatedly throughout this is, do similarly situated firms pay the same rates? The answer to that is yes. And that is in fact the purpose of the Exhibit A process. There's a comment down below, "nor can we understand or evaluate the full amount of revenues earned by the exchange." I'll just mention that. I won't respond to it now. It's mentioned several times. I'll respond to it in a moment. Question two, I think I largely answered, is, what do different classes of market participants pay? The answer is, with one exception, which we've already talked about, there's no such thing as different classes of market participants. If you are using the data in a certain way, it matters not to Network A that you are a broker, that you are a mutual fund, that you are a news organization. Network A does not distinguish by type of customer, it distinguishes by how, and to what extent, and what data will you be using, with one exception being the nonprofessional that we have just discussed. It's the only distinction as between so-called classes of market data users. And as I said a moment ago, certainly they have sliding scales. Tom showed them to you. Rates are not Diversified Reporting Services, Inc. (202) 296-9626 47 different, fees can differ, certainly, according to different uses and according to size. I actually may have skipped ahead for the moment. There's just more about market participants being treated differently for no specific reason. I simply would respond that it's all in the fee schedule, and the fees are applied uniformly, and the schedules are all standard. How are specific arrangements negotiated? What's the relationship to various exchange rebate programs? I think the operative word there is, "negotiated," and when you look down, it talks about the network administrator, what terms they negotiate, what discretion they have. Tom has largely covered this. The Exhibit A and the contracts are printed, standard forms. There are no terms to be negotiated. There is no discretion allowed vis-a-vis the network administrator. And probably the most -- an example I can point to -- and perhaps Bob Colby will remember this. It was the one pilot that exists today, the CTA pilot. About four years ago, we came down to the SEC, because one of the first things that, frankly, I decided, or wanted, to do when I inherited market data four years ago, was take real-time -- take delayed data off of television. I didn't much care at the moment whether we replaced it with real-time, but I knew we wanted to get Diversified Reporting Services, Inc. (202) 296-9626 48 delayed off the television. It turned out, after we thought about it, there was no good reason why we couldn't put real-time on television. As we thought about that, we came down to the SEC, and we said, "There are two networks doing this, CNN, CNBC. Do you think this is a good idea?" They were very enthusiastic about the idea. We said, "Okay," then we would expect that we'll simply march up to their offices and negotiate something with them. The Commission's ultimate view on that was that that would be unacceptable, that we had to create a rate, publish a rate card, and administer -- offer and administer that rate uniformly to users, now up to five users. We were a little surprised by that, but nonetheless, it is what it is. And I think it illustrates the point that, at least as regards to the Network A administrator, there's absolutely no discretion and no authority whatever, even in that sort of silly example, to negotiate. The rest of this is about market data rebate programs. The New York Stock Exchange does not have such a program, so I will have to defer, to answer this question, to other markets who do have such a program. All the way down below, again to the extent that you can't read that, it continues to talk about negotiation. Hopefully, I have put that to bed. Diversified Reporting Services, Inc. (202) 296-9626 49 But there's a new concern raised. Concerns by market participants that they must, again, negotiate for market information with their governing self-regulatory organizations. A, there's no negotiation. B, it's a purchase of services. And I would suggest to you that broker/dealers purchase an endless number of services from self-regulatory organizations without any heightened degree of difficulty. The next one, I think, is on pilots, and Tom has already covered that. And so I won't even bother with that one, except to the extent that at the very end of that, if I have the right one -- yes -- at the very end of that, again, the discussion is largely about pilots there. Obtaining this information, i.e., the scope and the revenues associated with pilots -- and Tom perhaps didn't say it earlier -- the revenue associated with the cable TV pilot is de minimus relative to the overall revenues of CTA. But again, at the very bottom, "obtaining this information will also reveal the full amount of market information revenues earned by the exchanges." The full amount of market information revenues earned by the New York Stock Exchange is on line three of our annually published income statement. There is no -- maybe I misunderstood the point here. If the point -- if this is an accusation of fraud, Diversified Reporting Services, Inc. (202) 296-9626 50 financial fraud on the part of the New York Stock Exchange, the CTA, and PricewaterhouseCoopers, then, you know, perhaps we can talk about that. I didn't think that that's what was meant. Short of that, all revenues -- there are no double secret market data revenues, or any other kind of revenues that do not find their way into NYSE annually audited financial statements. How exactly does market data cross-subsidize other self-regulatory functions? There's a comment in there. Some have suggested that market information fees should support self-regulatory functions. The New York Stock Exchange has never made that suggestion. The New York Stock Exchange -- as I have said to this committee before, market data revenues are at a loss, and it is in no position to cross-subsidize any other function at the New York Stock Exchange. On the contrary, it is subsidized by other fees at the stock exchange. Down below, obtaining this information will help the committee understand how the exchanges and Nasdaq account for market information costs and revenues. I can help you there. We account for market information costs and revenues on an accrual basis, in accordance with generally accepted accounting principles. The last question. Mercifully the last question. Systems maintenance costs, administrative costs exchanges Diversified Reporting Services, Inc. (202) 296-9626 51 incur for aggregating and disseminating market data. If it isn't obvious to most people around this table, we do not produce cost accounting for -- in our financial statements for market data or, frankly, any other business line at the stock exchange, for the simple reason that there is no generally accepted accounting principle framework in order that would enable us to do that. There is no great difficulty in producing a cost accounting P&L for -- well, there is a fair amount of difficulty, but it can be done for market data. That's not the issue. The issue is that you could produce a hundred P&Ls for NYSE market data, and no one would be better than the other, because there would be very substantial and, frankly, arcane assumptions as to the categories of fees that are appropriately used, and the percentage of allocation within those categories that -- I think I've said before, I think whatever debate exists on market data today would pale in comparison to the debate of producing a cost accounting P&L for market data. So I will draw this to a merciful close. The punch line to all of this is that all rates are established through consensus among the broker/dealer community, and indeed through a public process via the SEC. All rates contracts are public and standard, uniformly applied. Contract administration is designed to ensure uniformity. Diversified Reporting Services, Inc. (202) 296-9626 52 The NYSE, the CTA, and other participating markets, annually produce audited financial statements with their P&L. It's not clear what more we could do vis-a-vis transparency of the process but, frankly, we're open to suggestions. And I would suggest that none of that is surprising -- uniformity, standardization -- given the governance structure that the New York and the other markets live under. So I'll stop there, Joel. DEAN SELIGMAN: Okay. Are there questions? Let's start with George. Then I saw Eric. I wasn't looking at this side of the table. MR. JENNISON: Bob, I'm curious. In answer to the second to the last question, you said that -- you seemed very confident in saying that the market data operates, and the New York Stock Exchange operates at a loss, and is in fact subsidized by other areas. And then in answer to the last question, you said that, you know, talking about costs, and costing out the entire operation, it would be difficult to do, but could be done, and you hadn't really done -- MR. BRITZ: No, I didn't say we hadn't really done it. And if I said, "difficult to do," I only meant to the extent that it would require assumptions and allocations. Not difficult to do, George, difficult to agree upon the underlying allocations and assumptions that are in there. But I'll answer the first part of your question, if Diversified Reporting Services, Inc. (202) 296-9626 53 I may, which is that I am not smart enough to know, when a trade takes place, are we executing to investors' orders, or are we producing the price. And clearly, we're doing both. But if you're going to get into a cost-accounting exercise, which, as you all know, is more art than science, you have to say, "Well, a percentage of this box is dedicated to creating that price, and a percentage of this other box is dedicated to actually executing investors' orders." My own personal theory, which you've heard before, is that it's all about the price, because if we were in the order-execution business, we could e-mail the counter parties as to the details of their trade, and that would be perfectly okay with them. But there are probably millions of other people around the world who weren't party to that price, who have a very significant interest in that price, and more than a few sitting around the table. So the difficulty is in separating out those two, but I don't have to do that, frankly. I can look at transaction charges, I can look at market data fees, I can look at facilities. So that combining all of market operations, if you will, which includes all of that, it all operates at a significant loss. So I don't have to find angels on the head of pin in order to know that market data undoubtedly operates at a very -- tens of millions of dollars loss. Even my rusty cost accounting can deliver that Diversified Reporting Services, Inc. (202) 296-9626 54 conclusion very quickly. DEAN SELIGMAN: Eric? MR. ROITER: I think that was my question, as well. I thought I had heard you talk about two different types of accounting -- cost-based accounting and GAAP accounting -- but you're not making that distinction. I think you're saying that you can't really separate these closely interwoven threads, and that whatever set of assumptions you might apply in trying to quantify the costs associated with producing market data, those costs would exceed your revenues? MR. BRITZ: I actually made a broader statement, Eric, which is that the cost of executing investors' orders, and the cost of producing market data exceed the revenues from three revenue lines that contribute to that. So I guess what I have said is, there's no need, from my point of view, to drill down and try and divide the baby up. I know that all of market operations operates at a substantial loss. What I said as to the difference between cost and GAAP -- cost accounting is an internal management tool. There isn't an accounting firm in the world that will apply any certificate to a cost-accounting exercise, so it's not a surprise that we don't produce that in our annual financial reports. But certainly, we do look at cost accounting on an internal basis, as I suspect all of you do. Diversified Reporting Services, Inc. (202) 296-9626 55 DEAN SELIGMAN: Are there other questions? Carrie? MS. DWYER: I'm assuming -- maybe we ought to move on to the Nasdaq, because the comments that I have about all of this are more general and apply all of this. DEAN SELIGMAN: Before we do, let me just follow up with a couple of questions, and they're, in part, following up on George and Eric's point. If I understand what you're saying in response to, how is market data priced? You're describing a process. MR. BRITZ: Yes. DEAN SELIGMAN: To the extent one tries to understand the economic analysis, if you will, that goes to determining where fees should be, or how they should change over time, I don't have as clear a sense as to whether you do any economic analysis, and if so, what kind it is. MR. BRITZ: Well, I think you do, Joel. There are lots of variables that go into that. As I mentioned, in the last 10 years, they have all been on the rate-cut side, and probably amusing to this committee, but when I go to the finance committee of our board, and indeed the board as a whole, the biggest problem I have is explaining to them why this is a good idea to cut this particular fee by 50 percent, what the negative revenue implication is of that, and how we are either going to cut our own costs in order to offset that, or how we're going to recover that in some other fee. Diversified Reporting Services, Inc. (202) 296-9626 56 So the whole notion of how this proposal impacts non only New York, but I presume analyses are done in other marketplaces, as well, in terms of the ongoing funding and financing needs of the particular market is important. In terms of economic analysis, well, let's go back to the rate cut in 1997 by way of example. It was in the nonprofessional space. This was the per-quote usage-based fee. We had a program, and the over-the-counter NASD had a program, as well. We had 13 participants. All 13, to a firm, came to us and said that we had gotten it wrong. We did that -- by 1997, it was running about three years. We had gotten it wrong, and they much preferred the over-the-counter product. And we thought about that some, but at the end of the day, and a reasonable period of time, we said, "Fine, that's what they want, that's what we'll give them." It was a product that was in the marketplace, that was priced. We did nothing more than mirror the price and mirror the terms, because at the end of the day, that's what the customer said they wanted. The revenues were cut 50 percent in that example. That's not surprising that the customers wanted that. So you get to use benchmarks, sometimes, when you're lucky, that are out there in the marketplace. That's about the best way I know how to -- hopefully, I have given Diversified Reporting Services, Inc. (202) 296-9626 57 you some raw material for a reconciliations to that, Joel. DEAN SELIGMAN: All right. Now, to take it another level, there are differences between professional and nonprofessional fees. How do you determine where to set each? What's the economic analysis that you go through for that exercise? MR. BRITZ: Well, -- because economic analysis pretty much implies value pricing. And I can go back to that meeting on cable TV four or five years ago, but any number of other meetings, where it's been the SEC's point of view that, again, at least with regard to Network A -- that's all I know about -- that value pricing is not remotely part of the equation. Clearly, we don't all know what fair and reasonable means, and we have struggled with that over some period of time. And clearly, we're not in a cost-based metric. But the notion of applying economic value, of looking at a product that we will bring out, and saying that, "Well, McGraw Hill has a product similar to that, and they charge 50 bucks," is a license that the SEC has made very clear that we don't have. As regards the split between professional and nonprofessional, I would be candid to say I don't know. This was a little bit before my time, but I don't know that there was economic analysis. I know that there was an industry Diversified Reporting Services, Inc. (202) 296-9626 58 push to make real-time market data -- remember, at this time, the delayed product was pervasive in the nonprofessional space. That's a product that, for lots of obvious reasons, we didn't want to be associated with. It's an inferior product, it's not actionable, we didn't want our brand attached to it. It's very much behind why we wanted to take delayed off of cable TV. I don't know what viewers thought when they saw that billboard that said, "20 minutes delayed," but I know it couldn't have been anything very good. So the delayed product was pervasive, because at a hundred and sixty eight dollars, it was simply too expensive for most individual nonprofessional investors to have that. So when various vendors, and particularly broker/dealers, came to us, more in the late '80's, early '90's -- and frankly, Schwab was a pioneer in this category -- we sensed a consensus building that we needed to create a differential as between the professional price and the nonprofessional price, with the tradeoff being, or the benefit being, that we would make the data significantly more available. And that would inure to the benefit of the entire industry. DEAN SELIGMAN: Okay. I appreciate that. I think I've got a question from Andy. MR. BROOKS: So Bob, just to follow up on that, if Diversified Reporting Services, Inc. (202) 296-9626 59 in 1984 the rate -- there was one rate at a hundred and sixty four dollars, and now we have two rates -- a professional rate at $25, and a nonprofessional rate at $1 -- and there's virtually no difference between the data, because they're both real-time -- is that correct? MR. BRITZ: Real-time, yes, but there's a huge difference. MR. BROOKS: Well, I understand the point about how one professional could redistribute that data to their customer base, and call 50 clients and all that, and I appreciate that, but -- MR. BRITZ: No, there's even more of a difference in terms of the timing, the delivery system associated with that, whether it's inquiry-based versus streaming. There can be lots of differences. MR. BROOKS: Okay. MR. BRITZ: The similarity, Andy, is that they are both real-time. MR. BROOKS: Okay. Well, I guess I'd like to hear about this, maybe at some later point, but what I'm focusing on is the data that we're buying. The price has gone from a hundred and sixty five to twenty five, and the data that Charles Schwab or somebody else is getting has gone from a hundred and sixty five to a dollar. And I'm kind of thinking, I appreciate that we want to provide everything to Diversified Reporting Services, Inc. (202) 296-9626 60 everybody in America, and that's a wonderful thing, but again, have we -- are we being consistent in the application of these rates, and perhaps, are we disadvantaging the institutional/professional investor class, in order to cross-subsidize, or to benefit the individual investor? MR. BRITZ: Andy, your facts are slightly wrong. The data that goes from Schwab, or any other vendor, to a nonprofessional customer is a dollar a month, to be sure. The data that is displayed on Schwab's screen of a Schwab broker, a so-called professional, is -- we make no distinction. That's between broker/dealer and institutional. DEAN SELIGMAN: Bob, let me just follow up, though, on the response you made to Andy. It is worth clarifying. You suggested there are differences in the data, even though they're both real-time, that are going to professional and nonprofessionals. Could you particularize what those differences are? MR. BRITZ: Well, I'll give you one example. Professionals typically will have a market minder -- and some of the vendors can perhaps do a better job here than I can -- professionals typically have a market minder, which is a kind of a push technology. You don't need to constantly hit the enter key in order to get an update. Many nonprofessional products simply have a one shot, one bite at the apple every time you hit that key. That's just one example. Mark? Diversified Reporting Services, Inc. (202) 296-9626 61 MS. DWYER: But the exchange doesn't provide that. MR. MINISTER: Yeah, I was going to say. That is not exchange-differentiated, it's vendor-differentiated. We offer a streaming product of nonprofessionals also, so from an exchange standpoint, there isn't any difference, it's vendor-differentiated, I believe. MR. BRITZ: I think that's right. No, I think that's right. I guess what I was responding to was the product that the end customer receives can be very different than what you're looking at in terms of your professional -- trading desk, Andy. MR. BROOKS: Okay. Again, just to -- MR. BRITZ: But the point is a good one, everything that leaves SIAC on behalf of the CTA and the NYSE is real-time data. We are not in the delayed business. Everything that leaves us is uniform, that is true. DEAN SELIGMAN: So it's uniform to both professionals and nonprofessionals. They're getting -- MR. BRITZ: The pipe is the same pipe, the content is the same content. When it reaches professionals and nonprofessionals, Joel, it can look significantly different. DEAN SELIGMAN: But that's because of the intermediaries? The vendors, and so forth? MR. BRITZ: That is correct. DEAN SELIGMAN: Okay. Diversified Reporting Services, Inc. (202) 296-9626 62 MR. BROOKS: Just to follow up, if I can, if that's the case, our choice of vendors is our choice, and we're happy to use the vendor that's massaging this data in the way that we want it, but why should we be paying 25 times what the nonprofessional investor is paying to get the same raw data, if in fact it's the same data? MR. BRITZ: Well, I mean, the simple answer to that, Andy, I think, is what we have been saying thus far. That the pricing process that we have described to you, in terms of grass roots at the individual broker/dealer level, and vendor level, and all of the vetting that that has done, has built a consensus around the notion that it is a good idea to increase the pervasiveness of the data, to discount off the professional rate for a nonprofessional customer. All of the SRO boards have endorsed that. It was publicly filed with the Securities and Exchange Commission. There was a public comment period, and the SEC ultimately approved it. So it's consensus-based. MS. DWYER: So Joel, just -- so this goes back to the comment I made earlier. So we're not talking about economic analysis, we're not talking about cost base, we're not talking about subsidy, we're simply talking about a pricing decision, based on what the market will bear, whether or not it's validated by the SRO boards, you know, which it is. But that's essentially, I think, the point. Diversified Reporting Services, Inc. (202) 296-9626 63 DEAN SELIGMAN: However one characterizes it, I think we've got it on the table exactly what the issue is. Bob, thank you very much. MR. BRITZ: Certainly. DEAN SELIGMAN: I'm going to suggest that -- let's take a 10-minute break, and begin again at ten to 11:00. We'll pick up with the Nasdaq. There is coffee right outside, and we'll just keep rolling. (A brief recess was taken.) DEAN SELIGMAN: We're going to begin at this point with a presentation from Tom Davin and Nasdaq. MR. DAVIN: Great. Thank you. The Nasdaq stock market is pleased to be able to address the committee regarding our role as the administrator of the Nasdaq UTP tape plan. In response to Dean Seligman's letter that followed up on the March 1st meeting, we've put together a presentation that will address some questions that he raised in his letter regarding how we handle our role as the administrator. Basically, I've divided my presentation up into four main parts. The first is how market data fees are set. Next, the type of fee arrangements that exist, just the basic dollar figures. Third is contract administration -- how Nasdaq, as the UTP administrator, administers contracts with market data distributors. And then, lastly, just to touch on Diversified Reporting Services, Inc. (202) 296-9626 64 what the other responsibilities are of the administrator -- kind of a day in the life of the tape plan administrator, if you will. I think there's definitely some overlap in what I'm going to say and what's already been said by Bob and Tom. Wherever possible, I'll try to focus in on the areas where we're different, or where perhaps we're touching on some issues that weren't already touched upon, but there's obviously going to be some overlap here. Okay. To begin, I've got two slides regarding how fees are actually set. Basically, this is a sequence of events that we go through, as the tape plan administrator, and that the other members of our tape plan go through with us, as we change rates for existing fee structures, or as we create new fee structures for our market data. The first thing that we need to do -- or the first thing that happens is, we identify a business need or an opportunity. In theory, this could be something that sort of happens in a vacuum, just within Nasdaq. As the tape plan administrator, we could just come up with a great idea for a new fee structure, or a new level for an existing structure, and go ahead and push forward and try to implement it. In practice, it's been a lot different than that. I think, by and large, when we make changes or when we contemplate changes, it's at the behest, or as a result of Diversified Reporting Services, Inc. (202) 296-9626 65 requests made by the industry. It may be the online brokerage industry, it may be the market data vendors, it may be firms in the trading community. Typically, what we're looking for is proposals that are going to have application with more than one firm, that are going to have broad applicability, and they're going to benefit multiple firms. So once we've identified something that seems like a viable approach to go forward, we review it within the Nasdaq staff. We want to do some analysis on it, we want to compare it to our existing fee structures to make sure we've got consistency across the fee structures, that by implementing this new fee, we're not severely undercutting an existing fee structure unintentionally. Second, we want to take -- we do want to take a look and make sure there's broad applicability. We wouldn't want to respond to just one firm's concern, create a fee structure that would -- responded perfectly to their issues, but had no applicability to anybody else. We want something that is going to help the entire -- if not the entire industry, a broad spectrum of the industry, if we do make a change. And then finally, we look at the financial impact. We take a look at who the main customers might be of the -- or distributors might be under the pricing structure being contemplated, and we see what the impact would be on them. Diversified Reporting Services, Inc. (202) 296-9626 66 We take a look at existing distributors, and see what the financial impact would be on them. And then we look at our own revenues, and the tape plan revenues, and see what the impact would be of implementing the fee. As we're going through all these steps, we're talking to all of the potentially affected market data distributors, we're asking them, you know, what they think they might do, would they use a potential fee structure, how much usage might they have of it, and try to get a sense as to what the impact is going to be on our bottom line, and on the bottom lines of our customers. Next, if we feel as though we've got a viable fee structure, we take it to the Nasdaq committees and the boards that oversee our market data pricing. The board -- the Nasdaq board represents -- and Nasdaq members represent the general public, and also investors -- both individual investors and institutional investors -- but also represents Nasdaq issuers. All of these constituencies have a vested interest in Nasdaq market data pricing being reasonable and promoting broad distribution of the data. Obviously, as the users of the data who have to pay for it, the investors and the brokerage firms are very sensitive to the idea of a fee being higher than it needs to be. For Nasdaq issuers, they want their issues, the data on their issues, broadly distributed. Diversified Reporting Services, Inc. (202) 296-9626 67 They wouldn't want overly high Nasdaq pricing to retard that function. So the representation of our board would certainly keep in check any tendency we might have to overcharge for the data. The next step, if it's approved by the Nasdaq