Comments of Charles Schwab & Co. Inc. on SR-NYSE-2001-42

November 20, 2001

Via Electronic Delivery

Jonathan G. Katz
Secretary
Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, D.C. 20549-0609

Re: Notice of Filing of Proposed Rule Change by the New York Stock
Exchange, Inc., Establishing the Fees for NYSE OpenBookTM
Release No. 34-44962/File No. SR-NYSE-2001-42 (the "Release")

Dear Mr. Katz:

Charles Schwab & Co., Inc. ("Schwab") appreciates the opportunity to comment on the NYSE's proposal to establish fees for its new OpenBook service which will permit subscribers to view limit orders contained in the NYSE's limit order book. Schwab fully supports making this important depth-of-book information available to all investors as soon as possible. However, because the current pricing structure for OpenBook will effectively preclude retail investors from viewing their own orders and the orders with which their orders compete for executions, we do not believe the proposal as currently drafted can be approved consistent with the requirements of the Exchange Act.

In our view, both the fee structure and the restrictions on how the OrderBook data can be used are unreasonable and unfairly discriminate against individual retail investors. The NYSE's statement suggests that there is a need to control the launch of OpenBook until technical issues are resolved, and we have no reason to question that as a concern. Nonetheless, the ultimate goal of the proposal should be to ensure that depth-of-book information is distributed on fair and reasonable terms that allow all market participants and investors to use it as they believe appropriate. Accordingly, we believe the Commission should deny the proposal as drafted and encourage the NYSE to work with its members and other market participants to revise the proposal so that OrderBook information will be made available on a fair, non-discriminatory, and procompetitive basis as soon as practicable.

Summary of the Proposal

The NYSE is proposing two sets of fees for its OpenBook service. First, it intends to charge each entity that receives the OpenBook data feed a fee equal to $5,000 per month. Second, it intends to charge an end-user fee of $50 per month for each terminal through which the end user is able to display the OpenBook. Thus, as we understand it, a broker-dealer would pay both an annual fee of $60,000 to receive the OpenBook data feed and an annual fee of $600 for every internal computer terminal on which OpenBook can be displayed. In addition, if a broker-dealer or other "vendor" makes the OpenBook display directly available to its customers, each such customer would incur (or result in the broker-dealer or vendor incurring) an annual fee of $600.

The NYSE intends to impose two restrictions on the use of OpenBook. First, the NYSE will initially prohibit any recipient of the OpenBook data feed from redisseminating it in any form. In other words, anyone wishing to receive the data feed may receive it from the NYSE only.¹ Second, recipients of the data feed who externally redistribute the OpenBook display must do so in the window format prescribed by the NYSE. Vendors and broker-dealers will be prohibited from making any enhancements to the content or format of the display or consolidating the data with other market data (including better-priced orders available in other markets) to create a more useful product for end users such as retail customers.² By contrast, institutional investors, market professionals, and broker-dealers who use the data internally will be allowed to make any changes they wish, including consolidating the display with other data and integrating the information into data analytics, trading models, and order-management systems.

Comments

Real-time market data is critical to the efficient and fair operation of the capital markets and to the ability of investors to determine the best prices available in the markets and make informed trading decisions. Individual investors have traditionally been at a disadvantage to institutional investors and market professionals in terms of the availability of such data. While institutions and professionals have had cheap, real-time access to market data, retail investors have had to rely on delayed quotes or - when broker-dealers and vendors have been willing and able to pay duplicate fees - on query-based (non-streaming) real-time information. With rare exception, such data is available only one NBBO quote at a time. While some progress has been made over the past several years in leveling the playing field for retail investors, they continue to have far less access to useful information than institutions and professionals.

This informational disadvantage has become more pronounced with the advent of decimalization, which has made top-of-the book data - the only data exchanges are currently required to disseminate - less and less meaningful. As the SEC's own data has demonstrated, in the post-decimal environment, quote depth at the inside market has declined by 60-80% in both the exchange-listed and Nasdaq markets. Moreover, in a post-decimal world, traders may step ahead of a retail investor's limit order for only a penny a share (or even less in markets that allow sub-penny pricing). This fact makes it particularly critical for retail investors to be able to see existing limit orders (and to be able to modify their own limit orders) away from the inside quote. In an environment where the inside quote changes rapidly, where most trading interest resides below the top-of-the-book, and where traders can step ahead of the best-quoted price for a penny, access to depth-of-book information has become essential to determining the state of the markets and making fully informed trading decisions.

While the NYSE proposes to make such depth-of-book information available, both the fee structure and the restrictions on redistribution deprive retail investors of equal and fair access to the same type of information institutions and professionals will enjoy. Broker-dealers will have two options for making the OpenBook information available to their retail customers. First, they can make the display available internally to registered representatives who can provide the information to customers who telephone and request it. To do so, however, they must pay a per-terminal fee of $600 a year. Thus, for example, a firm with 5000 registered representatives who serve retail customers would be required to pay the NYSE an annual fee of $3 million to make OpenBook information available on their screens. This cost would make widespread internal dissemination within firms serving retail customers prohibitively expensive.

The far more effective way for a broker-dealer to make OpenBook information available to its retail customers - and the far more meaningful way for customers to receive it - would be directly through electronic or online access. To do so, however, the firm would incur a per customer end-user fee of $600 a year. Thus, for example, Schwab would be required to pay the NYSE an annual fee of $300 million just to make OpenBook information available to half a million of our online customers, a small percentage of our total online accounts. It is unrealistic - and unjustifiable - to expect a firm to absorb this kind of expense³, and it is unfair and discriminatory to impose such a cost on online retail investors. The NYSE's statement does not justify or attempt to explain the reasonableness of the $50 per device or use fee in light of the data's current monopoly status.⁴

Finally, even if the cost disincentive could be overcome, the restrictions on the form and content in which OpenBook can be redistributed to retail investors would provide them with a far inferior information product than would be available to institutions and professionals. As stated above, broker-dealers and market data vendors would be required to redistribute the information in the precise format dictated by the NYSE and would be prohibited from enhancing it, making it more useful to their customers, or consolidating it with other market data. Thus, for their $600 a year, retail investors would receive a plain vanilla, one-size-fits-all information product, while for the same price, institutional investors and market professionals would be able to integrate the data into any number of value-added information products. Moreover, the format and redistribution limitations will provide the NYSE with an unfair competitive advantage as a market data vendor, thus allowing it to capitalize on its SRO status.⁵

The NYSE's statement suggests that it believes there is currently no retail demand for this data. As contemplated by the Exchange Act, the information should be made available to all investors equally. The extent of retail demand should be determined by market forces and open competition, not arbitrarily by the NYSE. As we have stated elsewhere, market data does not belong to the exchanges, but to the individual investors and broker-dealers who create it and are required by regulation to give it to the exchanges for free. Distribution of this critical data must meet the Exchange Act's requirement that the proposal not unnecessarily discriminate among market participants and not unduly burden competition. Without a cost-effective alternative for retail investors, the OpenBook proposal does not meet these critical investor-protection standards.⁶

Conclusion

The NYSE's proposal illustrates most of what is wrong with the current market data system and highlights the need for reform. Schwab and a group of other firms have submitted, in our separate statement to the Report of the Advisory Committee on Market Information, has outlined ways in which the market data system could be effectively reformed to better serve the public interest. However, the Commission need not address these larger issues in determining whether to approve the NYSE's OpenBook proposal. The proposal on its face unfairly discriminates against retail investors and unduly restricts the availability of critically important market data on a fair and equal basis. As such it fails to comply with the Exchange Act and should be denied.

We appreciate the Commission's consideration of our views on this matter and would be happy to provide any additional information the staff might find useful.

Very truly yours,

W. Hardy Callcott
Senior Vice President and
General Counsel

cc: Harvey L. Pitt, Chairman
Isaac C. Hunt, Jr., Commissioner
Laura S. Unger, Commissioner
Annette Nazareth, Director, Division of Market Regulation

Footnotes

¹ The Release itself is somewhat unclear on whether the initial restriction applies only to redissemination of the data feed itself or whether it acts as a complete ban on external redistribution of the OrderBook display. Information on the NYSE's OpenBook web site (www.nysedata.com/openbook) indicates that the initial prohibition applies only to redissemination of the data feed itself. As stated thereon under the caption Vendor Implementation: "Vendors may distribute only display services to their external customers; no data feed redistribution is permitted. Vendors that externally redistribute NYSE OpenBook may not commingle NYSE OpenBook with any other order information from any other market." Should the SEC approve the NYSE proposal, the approving Release should clarify precisely how these restrictions will apply.
² On its face, this violates the Display Rule, Rule 11Ac1-2. The NYSE, when it chooses to act as a market data vendor, must follow the rules that apply to other market data vendors.
³ It would also place retail firms, in particular firms with a large online retail client base, at an unfair competitive disadvantage to firms that cater to institutional investors or serve their clients solely through telephone and in-person service.
⁴ We note that Nasdaq's fee for streaming Level II data for "nonprofessionals" - while still high - is at a slightly more reasonable price of $10 a month, reflecting the need for a different fee structure for retail investors.
⁵ The NYSE's statement suggests that this distinction is necessitated by difficulty in policing restrictions on internal use. We are confident, however, that the NYSE could find ways to adequately audit how the information is being used internally. More important, this purported difficulty is insufficient to make the proposed discriminatory and unfair terms acceptable under the Exchange Act.
⁶ There is a significant threshold issue whether data gathered by an SRO pursuant to statutory requirements that are binding on all market participants can or should be sold outside of a National Market System Plan without sufficient safeguards and checks and balances in place. This is an issue that the SEC's market data concept release and the advisory committee on market information touched upon, but did not resolve. Such issues should not be resolved through an SRO rule filing approval with only 21 days notice to consider the implications. We note that the NYSE did not seek comment on the OpenBook proposal from its members before submitting it to the Commission for approval, thereby belying the Exchange's claim that its internal processes when considering market data issues are sufficient to protect the interests of individual investors.

¹	The Release itself is somewhat unclear on whether the initial restriction applies only to redissemination of the data feed itself or whether it acts as a complete ban on external redistribution of the OrderBook display. Information on the NYSE's OpenBook web site (www.nysedata.com/openbook) indicates that the initial prohibition applies only to redissemination of the data feed itself. As stated thereon under the caption Vendor Implementation: "Vendors may distribute only display services to their external customers; no data feed redistribution is permitted. Vendors that externally redistribute NYSE OpenBook may not commingle NYSE OpenBook with any other order information from any other market." Should the SEC approve the NYSE proposal, the approving Release should clarify precisely how these restrictions will apply.
²	On its face, this violates the Display Rule, Rule 11Ac1-2. The NYSE, when it chooses to act as a market data vendor, must follow the rules that apply to other market data vendors.
³	It would also place retail firms, in particular firms with a large online retail client base, at an unfair competitive disadvantage to firms that cater to institutional investors or serve their clients solely through telephone and in-person service.
⁴	We note that Nasdaq's fee for streaming Level II data for "nonprofessionals" - while still high - is at a slightly more reasonable price of $10 a month, reflecting the need for a different fee structure for retail investors.
⁵	The NYSE's statement suggests that this distinction is necessitated by difficulty in policing restrictions on internal use. We are confident, however, that the NYSE could find ways to adequately audit how the information is being used internally. More important, this purported difficulty is insufficient to make the proposed discriminatory and unfair terms acceptable under the Exchange Act.
⁶	There is a significant threshold issue whether data gathered by an SRO pursuant to statutory requirements that are binding on all market participants can or should be sold outside of a National Market System Plan without sufficient safeguards and checks and balances in place. This is an issue that the SEC's market data concept release and the advisory committee on market information touched upon, but did not resolve. Such issues should not be resolved through an SRO rule filing approval with only 21 days notice to consider the implications. We note that the NYSE did not seek comment on the OpenBook proposal from its members before submitting it to the Commission for approval, thereby belying the Exchange's claim that its internal processes when considering market data issues are sufficient to protect the interests of individual investors.