October 10, 1997

Jonathan G. Katz, Esq.


Securities and Exchange Commission

450 Fifth Street, NW

Washington, DC 20549

Re: SEC Release No. 34-38672; File No. S7-16-97

Dear Mr. Katz:

The National Association of Securities Dealers, Inc. ("NASD"), and its subsidiaries, NASD Regulation, Inc. ("NASD Regulation") and The Nasdaq Stock Market, Inc. ("Nasdaq"), are responding to the request for comment in the Securities and Exchange Commission ("Commission" or "SEC") Concept Release regarding the reevaluation of the SEC’s approach to the regulation of exchanges and other markets because of technological advances and the growth of alternative trading systems ("ATSs"). The NASD and its subsidiaries commend the Commission and its staff for analyzing the complicated issues that technological advances and investor demands raise with respect to the existing U.S. regulatory framework for regulating trading markets.

The Concept Release discussion of possible approaches to reevaluating its approach to regulation is part of a long line of initiatives on market structure that the Commission and the self-regulatory organizations ("SROs") have taken over the last twenty-five years. We see the discussion found in the Concept Release as a direct outgrowth of the valuable steps that the SEC has already taken in promulgating the Order Handling Rules in 1996. 1 To date, these new Rules have provided significant benefits for investors. The Concept Release provides an analytical framework for considering the next steps necessary to respond to developing trends reflecting the ever changing effects of technology on the mechanisms of our securities markets. As the process moves forward beyond the concept stage, we believe that SROs, investors, and the securities industry should be given additional opportunities to analyze clearly defined approaches to ensure that our robust capital markets are not harmed and that individual investors continue to be afforded the protections that the SEC has diligently sought to maintain. 2

In evaluating possible changes to the Commission’s historical framework for regulating markets, the paramount consideration must be the protection of investors. As discussed more fully below, our strongly held view is that regulating most ATSs as broker-dealers has worked well to address the relevant regulatory concerns identified by the Commission and that broker-dealer regulation can be enhanced in ways suggested by the Commission to promote the goals of investor protection and preservation of robust capital markets. In our view, the new concept of a tiered exchange approach to ATSs creates a significant risk that the existing investor protection standards may be undermined, and innovation discouraged, because this approach may fragment the existing cohesive regulation of the equity markets and dramatically increase the costs to the market without providing significant added benefits. Moreover, the tiered exchange approach may, depending on the unspecified details of implementation, impose unnecessary burdens on competition between traditional exchanges and ATSs.

While we believe that the domestic regulation of ATSs is relatively well developed and can be enhanced within the existing framework, the area that deserves greater attention is the regulatory structure surrounding U.S. investors’ access to foreign markets and foreign markets’ access to U.S. investors. These are complex and critical issues that need the leadership of the SEC, in conjunction with the SROs, investors, and the securities industry, to ensure the protection of investors and the maintenance of fair and orderly markets. Whatever regulatory approach is selected to address these developments must ensure that U.S. regulators and self-regulators have the necessary access to information and enforcement tools to protect U.S. investors and the integrity of U.S. markets. In addition, any approach must ensure competitive fairness to U.S. markets as they extend their reach to foreign markets. While the approaches to regulation of foreign markets’ access to U.S. investors outlined in the Release merit consideration, we believe that more study is required before the proposal of a new regulatory framework.

I. Overview

The Securities Exchange Act of 1934 ("Exchange Act") reflects the Congressional mandate that the securities laws and rules should ensure, among other things: (1) the maintenance of fair and honest securities markets; (2) the protection of investors operating in such markets; and (3) the development and improvement of the mechanisms of a national market system for securities. Under the Exchange Act, the SEC and the SROs, having due regard for the public interest, the protection of investors, and the maintenance of fair and orderly markets, have fostered multiple improvements in our securities markets, guided by Congress’ determination that the national markets must permit:

< economically efficient executions;

< fair competition among brokers and dealers, among exchange markets, and between exchange markets and markets other than exchange markets;

< public availability of quotation and transaction information;

< an opportunity to obtain best execution; and

< an opportunity to obtain execution without dealer intervention, to the extent that such execution is consistent with economically efficient executions and the opportunity to obtain best execution. 3

Responding to these Congressionally mandated goals, the Commission and the SROs already have put in place a regulatory structure that is well designed to address the basic regulatory concerns identified by the Commission in the Concept Release. With the implementation of the Order Handling Rules this year, the existing regulatory structure addresses the most significant transparency and access concerns and appears to work well to address the Congressional mandate. The Concept Release does not provide meaningful data on why this framework should be abandoned.

Of equal importance is the unified self-regulatory approach that has prevailed since the enactment of the Exchange Act in 1934, when exchanges, such as the New York Stock Exchange ("NYSE") and the American Stock Exchange ("AMEX"), were formally registered as statutorily recognized SROs, and its subsequent amendment in 1938, which led to the creation of the NASD as an SRO with respect to equity and debt securities not listed on an exchange. This approach has relied upon the primary markets for equity securities to play a vital role in the oversight of trading activity in the securities markets and develop the extensive mechanisms required to meet the statutory mandates of fair and orderly markets, investor protection, and perfecting a national market. For example, for all Nasdaq and non-Nasdaq equities and debt securities, there is in place a consolidated self-regulatory examination, surveillance, investigation, and disciplinary system, now implemented via NASD Regulation. In addition, the NASD has recently undergone significant governance reform to enhance and ensure the independence of our regulatory efforts from our market operation and other commercial functions. We believe these reforms are well designed to address the consolidated audit trail, conflict of interest, and other concerns identified in the Concept Release.

Finally, we believe that ATSs will proliferate, participation in the market by individual investors will likely continue to grow, and methods of communication (most notably, the Internet) will become more sophisticated and less costly As a result, as is already the case in some instances, price discovery and leadership will not be consistently focused on one or even a few markets or points of execution. This development, we believe, is largely beneficial for markets as it contributes to innovation and market efficiency, and market fragmentation issues can be and have been addressed under the existing framework for regulation. From a regulatory point of view, this development underscores the importance of the unified self-regulatory system within which the markets and ATSs currently operate.

II. Methodology of Analysis

We agree with the Commission that, with the National Market System goals as a starting point, the best approach to developing concepts and, ultimately, specific regulations for new regulatory structures is to define the principles that should govern market structure and regulatory structure choices. One can then analyze the two basic approaches proposed in concept and determine which of these two proposed approaches best meet such principles. 4

III. The Key Principles - Domestic Markets

We believe the Commission has correctly articulated several important general principles that should form an analysis of market structure and its regulation. Specifically, in our reading of the Release, we identified the following principles that appear to govern the SEC’s approach to ATSs and exchange regulation:

. The Best Market Structure Provides High Levels of Transparency and Efficient Means of Access. Market participants making trading decisions must have a real-time view of all relevant trades, order, and quotes, and the ability to trade against all visible quotes and orders. This principle ties closely to the statutory goals regarding investor protection and perfection of a national market system.

. Innovation Must Be Encouraged. Market structure should be designed to encourage innovation and choice among multiple markets and trading mechanisms. Governmental intervention into market structures should be minimized to the extent possible, because such intervention tends to impede innovation.

. Market Regulators Must Have Access to All Relevant Trading Information. Regulators and self-regulators must have a consolidated view of market conditions and events, including historical trades -- trade reports -- and prospective or proposed trades (orders and quotes), whether such information comes from participants in organized markets or private trading systems. In addition to possessing basic audit trail information, regulators and self-regulators must be able to obtain quickly answers to questions needed to supplement the audit trail information.

. Regulation Should Address Conflicts of Interest Between An SRO’s Market and Regulation Operations. In order to ensure rapid responses to requests for information necessary to supplement audit trail information regularly possessed by self-regulatory organizations, firms and individuals subject to an SRO’s jurisdiction should not have any cause to question the SRO’s integrity and freedom from competitive conflict of interest.

While these principles are relevant to any analysis of the regulatory approach to exchanges and ATSs, we believe that there are other, equally important principles that should be applied to any analysis regarding the development and overall design of market regulation, but did not appear to be as clearly stated in the Concept Release as those mentioned above:

. Unified Self-Regulation Promotes Investor Protection; Fragmenting This System Would Threaten to Decrease Regulatory Efficiency. The existing self-regulation structure is actually unified in the case of Nasdaq and non-Nasdaq unlisted securities and effectively unified in the case of exchange traded securities where a primary market tends to act as the primary self-regulator. This unified approach has been developed over time by Congress, the Commission, and the securities industry to ensure that investors that trade securities in U.S. equity markets are protected from illegal activity and unethical business practices. The introduction of multiple SROs, especially those that may be subject to scaled back regulatory standards, poses a serious risk for an increased potential for gaps in regulation or differential standards of examination, surveillance, and enforcement. The resulting fragmentation diminishes the ability of the regulatory structure to detect and deter illegal and unethical behavior. Moreover, the costs of regulation will increase as the inefficiencies that additional SROs cause increase.

. To Promote Fair Competition, Competing Markets Should Operate Under Comparable Regulation. Entities that are going to compete with each other for trading activity should be comparably regulated. If an entity seeks to establish itself as an exchange, it should be subject to comparable restrictions and obligations as other exchanges. Otherwise, it should not be entitled to call itself an exchange. Differential standards for regulated markets also pose the potential for substandard market surveillance, examination, and enforcement efforts.

IV. Analysis

The SEC has identified two possible means to regulate the domestic markets in the years ahead: "enhanced broker-dealer regulation" and "flexible exchange regulation." Our analysis of how these two respective approaches satisfy the principles outlined above follows.

A. Enhanced Broker-Dealer Regulation

Regulation of ATSs as broker-dealers is the approach that is currently in place and has worked reasonably well to date to achieve appropriate levels of transparency, competition, access, and oversight. ATSs are broker-dealers that act as new providers of liquidity, i.e., they provide an additional vehicle for liquidity through displaying orders to market participants and providing trade execution and matching services. Traditional market makers and specialists are categories of liquidity providers that have operated for years within organized markets. A market maker is a broker-dealer that displays its own quotes and others’ orders, and acts as a principal to provide liquidity; a specialist is a broker-dealer that displays its own quotes and others’ orders and provides liquidity to a market by acting in some cases as a dealer and in others as an agent; and ATSs are simply a new form of broker-dealer liquidity provider (analogous to a brokers’ broker) that acts strictly as agent. Each liquidity provider, whether a market maker, a specialist, or an ATS, successfully operates within markets that are self-regulatory organizations. We believe that enhancement of this approach can meet all of the market structure principles as well as the regulatory structure principles, set forth above.

From a market structure perspective, transparency of, and fair access to, price information is a key principle that must be preserved and even enhanced. Thus, in any new structure that is contemplated, equity securities order and quote information that is simultaneously made available to a significant subset of investors and market participants should be consolidated and widely disseminated to all investors and other market participants through a centralized display system. Current SEC and NASD Rules have significantly increased the amount of information from ATSs that is displayed in Nasdaq. For example, with the promulgation of Exchange Act Rule 11Ac1-4 to require the display of customer limit orders in market maker and specialist quotations and the amendment of Exchange Act Rule 11Ac1-1 to require the display of market maker and specialist price information contained in ECNs, together with Nasdaq’s development of a linkage mechanism to display and access ECN best-priced orders, the transparency of the markets, and especially the Nasdaq market, have been greatly enhanced. As a result, spreads in Nasdaq securities have narrowed dramatically without any identifiable loss of liquidity, thereby benefitting investors.

Further, through Nasdaq’s provision of a trading link to ECNs, investor access to previously inaccessible ECN price information has increased dramatically, at a fairly limited cost. Thus, in our view, the broker-dealer framework for accommodating market structure reform has worked very well and has successfully met the principles outlined above. Market transparency has improved, and innovation has been encouraged (both in terms of the number of ECNs that are operating and in connection with the way that the primary over-the-counter market--Nasdaq--links to these ECNs). Further, because certain ECN order information that previously was not captured real-time is now included in our market data, NASD Regulation has more efficient access to trading information. To the extent that additional data is needed, we believe that the existing framework provides sufficient flexibility to require such data. Moreover, regulatory fragmentation has been avoided because a single SRO, NASD Regulation has responsibility for overseeing all of the trading that occurs through these linked systems operated by NASD members.

Even with these new rules, certain issues relating to the existence of a "two-tier market" continue. As the Commission recognized in the Order Handling Rules Adopting Release, the new rules did not require ECNs to display information from institutions and other non-market makers in the public quote system. 5 We believe that this failure to capture this information leaves a gap in the current regime that should be addressed promptly to meet the statutory goals of transparency and the elimination of two-tier markets. In our view, transparency could be improved further by requiring that this non-market maker order information be included in the publicly disseminated best bid and offer quotation either through SRO requirements or a revision to the definition of a "qualifying ECN" in Exchange Act Rule 11Ac1-1. This enhancement to the existing structure could significantly increase transparency and minimize market fragmentation.

Of course, inclusion of this additional price information creates other complications that must be addressed contemporaneously with the enhancements to the price information requirements, e.g., the problem of locked and crossed markets caused by non-market makers and the access that these non-market makers would need to be able to access quotes and orders that their orders would lock or cross. We believe that these issues can be addressed through changes to SRO rules that clearly delineate the responsibilities of members and their customers when orders are displayed in the public quote system.

An enhanced broker-dealer approach also can address market access issues through enhanced order display and access rules (reaching, as suggested above, all orders, not just market maker and specialist orders). These rules allow access both in the sense of being able to see the best priced orders and in the sense of being able to trade against such orders. The only type of access not currently mandated is the ability to enter orders in a particular ATS. We are not aware of any evidence -- and the Concept Release does not describe any -- that denials of such access have been a problem. 6

Nonetheless, because the Commission appears concerned about denials of order entry access, we suggest that the Commission consider amending Exchange Act Rule 17a-23 to require a system sponsor to file with the Commission on a regular basis information concerning denials of access and to provide the terms and conditions on which such access denials are made. By enhancing its flow of information on access toATSs, the Commission would be better able to ascertain whether there is a denial of access issue that must be addressed through a more stringent approach.

If the Commission discovers that denials of access in broker-dealer-regulated ATSs are a real problem, it would appear that the Commission may be required to seek expanded authority to regulate the problem. Under previously proposed and withdrawn Rule 15c2-10, the SEC would have required the above-referenced information and created an appeal process, similar to SIP requirements, for denials of access by broker-dealer operators of ATSs. To the extent the Commission concludes both that it is necessary to regulate order entry access to ATSs and that an approach similar to the now-withdrawn proposed Rule 15c2-10 is not the preferable way to address this concern, we believe that the Commission should seek expanded jurisdiction over access to broker-dealer operated ATSs through legislative changes to the sections of the Exchange Act relating to exclusive securities information processors, Sections 11A(b) and 3(a)(22)(A) and (B) of the Exchange Act. Expanding the interpretation of the definition of the term "exchange" to include many private systems and creating an entirely new hierarchy of exempt, quasi-exempt, and non-exempt exchanges is not the means best designed to achieve this limited regulatory goal.

Under the regulation principles set out above, the enhanced broker-dealer framework also is well designed to provide an overall, consolidated view of the market to regulators and self-regulators via an integrated trade, quote, and order audit trail, and a rapid response to information not contained in the audit trail. With the contemplated implementation of the OATs 7 by NASD Regulation, information regularly available to self-regulators will include all historical and prospective trading interest. Moreover, members of existing SROs are accustomed to responding rapidly to requests for additional information and the existing SROs have been willing to use the compulsory authority to require information submission whenever a member has been slow to respond to appropriate requests for information.

B. Exchange Regulation

1. Our Understanding of the Concept

As outlined in the Release, the exchange regulation approach would involve three tiers of regulation for entities that the Commission would call "exchanges" under a new interpretation of the statutory definition found at Section 3(a)(1) of the Exchange Act. The lowest tier, Tier 1, would essentially exempt small and/or "low impact" markets from most exchange regulation requirements. Tier 2 would regulate larger, non-traditional exchanges as exchanges, but exempt them from regulatory requirements that conflict with their corporate structure and technological operations. Tier 3 would continue to regulate the traditional exchanges essentially as the statute currently contemplates. The tiered exchange approach starts with a premise that ATSs are different from ordinary broker-dealers, and through application of technology and development of new business practices, have started to perform functions more similar to exchanges. Hence, they should be treated as exchanges, if they (1) consolidate orders of multiple parties; and (2) provide a facility or set material conditions for system participants to agree to a trade. 8

The new interpretation of "exchange," if adopted, would be a reversal of the SEC’s current interpretation as enunciated in its Delta Government Options Corp. Release ("Delta"). Securities Exchange Act Release No. 27611 (January 12, 1990), 55 FR 1890 (January 19, 1990). We believe that the Delta analysis is the most effective interpretation of the term, and the SEC has not articulated a reason for revising this interpretation. Moreover, we do not believe that the new, proposed interpretation adequately addresses the issues and may unintentionally sweep into the exchange category entities that do not rise to the level of an exchange.

2. The Exchange Analysis

We agree with the SEC that ATSs are not typical broker-dealers and thus, may require some heightened form of regulation. Since 1989, when the SEC proposed Rule 15c2-10 to cover what were then known as proprietary trading systems, the NASD has supported the Commission’s proposals to provide ATSs with a special regulatory niche that allowed the ATS to compete and innovate, while affording investors and other participants in these systems the protections that they need. 9 While ATSs may require an added level of regulation to address certain issues not covered by ordinary broker-dealer regulation, we do not believe that these systems are exchanges. Rather, we believe that the Commission’s Delta discussion of what characterizes an "exchange" is correct. In Delta, the Commission stated that exchanges are systems "designed, whether through trading rules, operational procedures or business incentives, to centralize trading and provide buy and sell quotations on a regular or continuous basis so that purchasers and sellers have a reasonable expectation that they can regularly execute their orders at those price quotations." 10

As we analyze the statutory term, the Delta interpretation continues to have merit in determining which systems truly are generally understood by investors to be exchanges. Exchanges are trading venues where the investor has assurances that liquidity will be provided. The Delta interpretation ensures that any such systems are captured within the exchange definition, while excluding those that do not rise to the level of assuring liquidity on a regular or continuous basis. The newly proposed interpretation appears to capture episodic displays of price information where there is no assurance of regular or continuous liquidity. For instance, broker-dealer open limit order books provide no assurance that a customer will obtain an execution. Moreover, we believe that the new interpretation may cast too wide a net in that it would appear to capture market maker internal execution systems within its terms, as well as other traditional broker-dealer activities. The Commission has acknowledged in the Concept Release that one unanswered aspect of the exchange approach is that many of the systems that would be swept into the new tiered exchange program also continue to perform broker-dealer functions. Thus, the Commission has raised questions about the proper means to address the regulatory issues involved in regulating an entity that is both an exchange and a broker-dealer. 11 While it may be simple in concept to suggest that the NASD continue its oversight of the broker-dealer functions of an ATS, in fact the overlap between those functions and the exchange functions make the job extremely difficult, if not impossible. We do not believe that there are any simple answers to these questions, except to note that we believe that this is one of the most serious flaws in the tiered exchange approach: it captures entities that do not belong in the "exchange" box.

Given the difficulties in developing a new "exchange" definition that does not reach too broad a segment of the market and the relative ease in enhancing the current broker-dealer approach that flows from the Delta interpretation, we do not believe that the case has been made for developing a new interpretation. The Concept Release did not discuss in any detail the reasons for revising the long-held Delta interpretation. Unless the Commission is aware of additional bases or rationales for interpreting the term "exchange" differently, we see no reason to redefine such a critical term to create an entirely new regulatory approach, an approach that, as explained below, does not meet the key principles we posited above.

3. The Application of the Principles to the Exchange Approach

a. General Concerns

In assessing the desirability of changes to the existing regulatory framework for exchange regulation, it is important to assess how well a new approach would maintain effective surveillance and oversight of the markets and assure that self-regulators--the first line defense against unlawful and manipulative activity in the markets--are able to effectively police trading activity. In the Nasdaq Stock Market, where there is no traditional trading floor and trades are executed in a number of separate geographical locations, the NASD has developed an effective and unified surveillance program. All trades in Nasdaq securities--whether National Market or Small Cap--are reported to the NASD for dissemination to the investing public and for regulatory review. All quotations of market makers are similarly captured and subjected to sophisticated computer-based analysis to identify violations of NASD trading rules and possible manipulative activity. As mentioned above, those systems are in the process of being further enhanced by the development of OATs, which will capture additional information regarding the receipt of orders to enable the NASD to better police for best execution and compliance with the SEC’s order handling rules.

The development of ATSs as additional points of execution has been readily accom- modated within the existing regulatory framework. As broker-dealers, ATSs provide trade reports to the NASD for inclusion in the audit trail. Moreover, with the enactment of the order handling rules, ATSs are becoming linked to Nasdaq so that better priced trading interest of market makers is now captured in the NASD’s quotation surveillance data base.

We are concerned that the proposed tiered exchange approach is likely to promote a multiplicity of "ATS exchanges" and that such "ATS exchanges" will be subject to varying levels of regulation and oversight. Of particular concern is the possibility that multiple exchanges with varying levels of regulatory responsibility will create regulatory fragmentation that will be harmful to investors and will not promote fair markets.

For markets such as the Nasdaq Stock Market, where there is unified regulation and surveillance, introducing regulatory fragmentation would as a general matter threaten to create disparate enforcement standards (a regulatory race to the bottom), gaps in regulatory coverage, and increased costs resulting from the inability to continue to take advantage of economies of scale in developing and operating necessary enforcement, examination, and surveillance mechanisms and systems. In addition, we believe that the tiered exchange approach would subject regulated markets to an unfair competitive disadvantage while at the same time imposing unnecessary costs on ATSs, thus discouraging innovation by both fully regulated markets and ATSs. Our specific concerns are as follows.

b. Specific Concerns

1. Loss of Financial/Record-Keeping Oversight Capabilities

ATSs under the present regulatory environment are subject to broker-dealer financial and record-keeping requirements and are regularly examined by a self-regulatory organization. Given the risks entailed in the settlement services that are generally provided by ATSs, this level of financial regulation appears particularly appropriate. However, in the flexible, tiered exchange approach, none of these safeguards would be applicable. 12

2. Reduction in Effective Market Oversight Because of Regulatory Fragmentation

At a minimum, the existing surveillance systems will become less effective with the proliferation of new exchanges that are subject to varying levels of regulatory requirements and the resulting fragmentation of the necessary database that the primary regulator must rely on. As noted above, in the Nasdaq-trading environment, there is virtually no fragmentation of the audit trail today, because effectively all of the trading in such securities is captured in NASD Regulation databases. In the listed markets, each individual market center is responsible for policing activity in that market center. That system works effectively because of the significant price leadership role played by the primary exchanges--the NYSE and the AMEX--in the securities they list for trading, the existence of consolidated transaction reporting and audit trail systems, and extensive cooperation through the Intermarket Surveillance Group in providing protocols for investigating inter-market trading scenarios.

We are concerned that, in a tiered-exchange environment, existing audit trails will be made far less effective if those audit trails fragment through the creation of multiple "exchanges." The simple fact is that a surveillance audit trail can never be as effective if it captures only part of the trading activity. Any audit trail is supplemented by running transaction information through various parameters to identify suspicious incidents worth investigating. That process cannot be effective if that audit trail information is shared by a potentially large number of so-called exchanges. 13 As we mentioned earlier, NASD Regulation has expended considerable efforts in developing OATS to provide thorough scrutiny of order handling and execution. OATs could be seriously compromised by the tiered exchange approach and efforts to fix such problems will be costly.

We see no regulatory benefit in designating ATSs as exchanges if, as currently proposed in concept, the Commission apparently contemplates that certain "exchanges" may have limited or no real oversight responsibilities, except to maintain an electronic audit trail. Such an electronic audit trail is already present under the existing broker-dealer regulatory treatment of ATSs, and no change in regulatory structure is needed to establish one.

In addition, it is not clear under the tiered exchange approach whether it would be possible for a broker-dealer to limit its activity to that conducted through the ATS exchange and thereby avoid scrutiny by any full-fledged, or traditional SRO, such as the NASD or the NYSE. 14 Further, surveillance responsibilities may be adversely affected under the tiered-exchange approach; if any of the non-traditional "exchanges" lapses in its responsibilities, it will undermine the best efforts of the other exchanges.

3. Reduction in Effectiveness of Member Regulation

In order to conduct effectively examinations of member firms and their registered personnel to determine compliance with mark-up, suitability, and other important customer protection rules, access to integrated audit trails is essential. Thus, the fragmentation of regulatory information resulting from the proliferation of SROs will reduce the effectiveness of member regulation as well as market surveillance and regulation.

4. Regulatory Free-Riding

On a practical level, multiple SROs today share the duties of regulation. Under Section 17(d) of the Exchange Act, one SRO may contract with another SRO to have the second SRO take on responsibilities for the other SRO, and the Commission has the authority to relieve the first exchange of those responsibilities contracted away. The Commission appears to place great value in the likelihood that ATSs registered as exchanges will seek to contract with other SROs to handle their self-regulatory responsibilities under this section of the statute. This Section 17(d) arrangement, however, only works because each SRO, by statute, has the same responsibilities and obligations as any other SRO. In the proposed new framework, ATSs as "exchanges" do not appear to have equal regulatory obligations. Thus, the suggestion that these "exchanges" may be able to contract with other SROs to take on their regulatory responsibilities is unfair to the traditional SROs because it allows "free-riding" on NASD and NYSE broker-dealer regulation.

Moreover, it is not at all clear that a primary SRO performing regulation under contract with exempt or quasi-exempt SROs could expect and enforce rapid responses to requests for additional information from the contracting SROs and their members. Because one SRO would not have regulatory jurisdiction over another SRO or its members, the primary SROs remedies in the case of delay or outright refusal to cooperate would appear to be contractual, or to seek SEC intervention. Neither of these options seems particularly efficient, and both are certainly less efficient than the current unified approach to Nasdaq and non-Nasdaq unlisted securities..

5. Fair Competition Principles

The tiered exchange approach also would result in unfair competition among entities that are all entitled to call themselves "exchanges" but only some of which have all of the obligations imposed by the statute. Under the SEC’s approach, each tier of "self-regulators" would be subject to different levels of obligations regarding regulation of their markets and governance. In the Concept Release, the SEC provided no clear explanation of which obligations imposed by Congress are expendable and whether such obligations could be eliminated for all levels of exchanges.

By creating different levels of obligations, the tiered approach also is inherently unfair and misleading to the public because an entity can call itself an exchange while not having to live up to the same requirements under which other entities must operate. 15 Particularly troublesome issues related to this inequity are that: (1) certain tiers of exchanges would need few, if any, capabilities to regulate activity in their systems or obligations to supervise their participants; (2) the lower tiers would be able to expand more easily overseas than traditional exchanges can; (3) institutional access to their markets would be accomplished more easily than institutional access to traditional exchanges; and (4) the new types of exchanges would be permitted to have much simpler corporate governance standards that could allow a narrow band of interests to dominate and control their boards. Several of these factors run directly contrary to steps that the SEC has recently taken to provide greater investor protection and to ensure appropriate balance in exchange governance. Because lower tier exchanges under this approach would have a much easier regulatory burden, they obtain an unfair competitive advantage over the traditional markets.

6. Decrease Regulatory Efficiency/Increased Costs/Decreased Innovation

The tiered exchange approach also would appear reduce investor protection and market efficiency by limiting the economies of scale and scope available through one or a few primary SROs. The certainty costs of this approach also are large, as it relies upon hard-to-predict distinctions between "exchanges" and systems not exchanges, and between "exchanges" big enough to take on SRO functions and "exchanges" that would be exempt. These certainty costs would lead to reduced innovation.

In addition, the regulatory costs imposed on the traditional exchanges will likely stifle innovation, as more nimble, lesser-regulated entities that would be "exchanges" without the same costs could develop new systems and products with which even innovative traditional markets could not compete. Further, as we have noted above, it would appear that the government regulator would have to play an increased role in overseeing the new SROs that could be developed under this approach, with the result that the government’s oversight role could stifle innovation as well.

Moreover, the transition costs of moving to this approach are formidable, involving the development of appropriate SEC regulations, registration of all ATSs falling within the "exchange" definition, and the creation directly, or more likely by contract, of appropriate self-regulatory capacity by the "exchanges" falling in the non-exempt tier of exchange regulation.

Finally, the tiered approach to regulation creates inconsistencies with the notion that these ATSs are "private" trading systems; the tiered approach imposes on them virtual public utility regulation, but only if they happen to succeed. The imposition of such obligations will at a minimum impose substantial direct costs and will almost surely discourage the innovations private systems have historically brought to securities (as well as other) markets. Without more evidence of a problem to be solved, a more modest approach that focuses on perceived regulatory gaps seems more desirable than such a drastic change in the regulatory approach to successful private systems. 16

7. Conflicts of Interest Principle

Certain aspects of the Concept Release focus closely on the potential conflicts of interest between an ATS and an SRO that also operates a trading market. As noted in the SEC’s Release, when an SRO operates its own market, while regulating a broker-dealer member that operates a competing market, there are at least appearances of a conflict of interest. We believe, however, that the clear organizational separation of market operation and regulation within an SRO, as recently implemented by the NASD, combined with direct oversight by the SEC, is well designed to address the conflict of interest concerns raised in the SEC release. Under this structure, all enforcement decisions and requests for information are made by personnel completely independent of the operation of the market, and all information received in response to requests for information is available only to regulatory personnel for regulatory purposes. This structure should eliminate any hesitancy on the part of operators of private systems to provide information to NASD Regulation personnel and accordingly, we believe that the current broker-dealer approach to regulation of ATSs provides acceptable regulation without fear of conflicts of interest.

C. Conclusions Regarding ATS Regulation

Current market developments and technological opportunities present the Commission with a regulatory challenge. Multiple and diverse markets and trading mechanisms are beneficial to investors and indeed are required in order to efficiently support diverse trading needs and demands. Any concerns about market "fragmentation" can be, and historically have been, effectively and efficiently addressed by appropriate transparency regulations, by linking markets informationally and, most recently, by permitting access to all displayed orders across competing markets and ECNs.

Indeed, the recently adopted Order Handling Rules have established a valuable framework to address the issue of market fragmentation that has resulted after significant technological innovation created new capabilities for market participants to trade with each other. This framework created the means to ensure that buying and selling interests from a variety of sources are consolidated and disseminated to all market participants, and that those same participants can access them. We need more experience with the impact of the Order Handling Rules on ATSs before so fundamental a change as the Concept Release contemplates is proposed or implemented.

Unlike multiple markets and trading mechanisms, multiple regulators and self-regulators-- regulatory fragmentation--are not positive public goals and for Nasdaq and non-Nasdaq securities markets that currently enjoy unified self-regulation. Introducing such fragmentation would threaten to create disparate enforcement standards (a regulatory race to the bottom), gaps in regulatory coverage, and excessive costs due to the inability to take advantage of economies of scale in developing and operating necessary enforcement technology and mechanisms. Unlike market and trading mechanisms, developing the necessary links to address regulatory fragmentation is excessively expensive and, while potentially addressing informational concerns, does not in any event address concerns relating to differential enforcement standards.

For these reasons, while investor choice, competition, and linkage are and should remain the preferred approaches to the development and operation of trading mechanisms, consolidation of regulatory information and enforcement effort in one or a few truly independent regulators/self-regulators should be the preferred approach to regulation of these trading mechanisms and their users. In fact, the Order Handling Rules demonstrate that the broker-dealer model of ATS regulation works well to allow for competing trading mechanisms to operate while market fragmentation is minimized and the issue of regulatory fragmentation does not even arise.

Based on this overall view, we believe that an enhanced broker-dealer approach, i.e., an enhancement of the Order Handling Rules, supplemented with increased denial of access information under Exchange Act Rule 17a-23, is a better approach than a tiered exchange regulation approach.

V. Foreign Markets

While we believe that the domestic regulation of ATSs is relatively well developed and can be easily enhanced, the regulation of foreign markets accessible in the US is far less developed and threatens to be overrun by events if the Commission does not act in this area with some dispatch. For that reason we encourage the Commission to give priority to the development of specific proposals focused on this portion of the Release.

The regulation of foreign markets that choose to enter the U.S. electronically should be designed to eliminate any unnecessary barriers to U.S. investors participating in foreign markets. With technological barriers to cost efficient U.S. investor participation in foreign markets having fallen, any remaining unnecessary regulatory barriers should fall as well. While it is important to remove unnecessary barriers, investor protection requirements demand that the Commission maintain its regulatory and enforcement capabilities over U.S. investor activity that may occur through such electronic links. Thus, U.S. regulators and self-regulators must have the ability to collect information concerning trade, quote, and order entry activity in any system accessible in the U.S., and must have the capability to take enforcement action if activity is detected that harms investors or undermines the integrity of U.S. markets. Among other things, this requires that access be limited to systems whose home country regulators have formal information and surveillance sharing agreements with US regulators and self-regulators.

The Commission must also bear in mind the effect that its regulation may have on U.S. trading markets, systems, and issuers. In our view, the terms on which access is permitted in the U.S. to foreign systems must not put U.S. trading markets or issuers at an unfair competitive disadvantage. Competitive fairness should be pursued not only because it is per se appropriate, but also because the technological ease of establishing and operating off-shore markets could drive U.S. systems subject to unfair competitive disadvantage off-shore. In the case of disclosure regulation, any inequities could raise capital raising costs to U.S. issuers. In addition, consideration should be given to permitting favorable terms of access only to systems whose home country regulators provide reciprocal terms of access to U.S. systems.

While we believe this aspect of the Concept Release deserves the priority attention of the Commission, and while the approaches outlined by the Commission merit further consideration, we also believe the Commission must engage in more fact-finding in order to propose a more specific approach. Specifically, the Commission should develop more information concerning the various ways foreign systems can be made accessible to U.S. investors, the types of securities involved, the types of U.S. investors involved, and the types of U.S. intermediaries and information and systems vendors interested in this type of business, and the availability of reciprocal access to U.S. systems that might be interested in operating in other jurisdictions. The Commission could use this information to promulgate a more specific proposal and commenters in turn could provide insights both on the information the Commission collects and any specific proposal the Commission promulgates.

VI. Conclusion

In conclusion, the NASD welcomes the continuation of the dialogue to determine the best approaches for regulating the nation’s securities markets in the next century. Technology and investor needs and demands have changed some of the means by which investors trade. While the means of trading have and will continue to change, the ultimate ends of market regulation have not and will not greatly change: investor protection will remain as important as it has been for the last sixty years. We urge the Commission to continue to talk with all market participants on how best to achieve these core principles. We believe that these principles are best met by enhancing the current model and structure, as opposed to attempting to build an untested model that appears likely only to increase costs while potentially decreasing investor protection. Because the approach to foreign systems is far less developed than the domestic approach, we strongly encourage the Commission to devote its attention first to gathering the facts necessary to promulgate a more specific proposed approach to foreign systems.


Joan C. Conley


NASD, Nasdaq, and NASD Regulation


-[1]- Securities Exchange Act Release No. 37619A (September 6, 1996), 61 FR 48290 (September 12, 1996) ("Order Handling Rules Adopting Release"), adopting 17 CFR ß 240.11Ac1-4, and amending 17 CFR ß 240.11Ac1-1. These rules were adopted or amended by the SEC in September 1996 and went into effect on January 20, 1997. With respect to Nasdaq-listed securities, these rules will be completely phased in by mid-October 1997. Evidence to date has demonstrated that standard market quality measures, such as spreads, have dramatically improved for the securities that have become subject to these rules.

-[2]- Because of the difficulty in addressing precise issues at such an early stage of development of specific proposals regarding future regulation, we believe it would be premature to attempt to respond to all of the specific questions the Commission has posed in its Release.

-[3]- See Section 11A(a)(1) of the Exchange Act. This section was added to the Exchange Act in 1975. Pub. L. No. 94-29, 89 Stat. 97 (1975) ("1975 Amendments").

-[4]- We have chosen at this time not to address any specific points in the Concept Release to the extent that such points may relate to non-equity trading markets. We do note, however, that the approach outlined in the Concept Release for the regulation of systems that trade debt would cause even more significant changes to these markets than the proposed exchange-based regulatory approach would cause ATSs trading equities, without any evidence of regulatory need or investor protection concerns. We expect to comment on any specific proposals in this area the Commission may determine to publish for comment.

-[5]- Order Handling Rules Adopting Release, at 106.

-[6]- The information Broker-Dealer Trading Systems are required to maintain under Rule 17a-23 includes information concerning denials of access. If there were any occurrences of such denials, we assume the Commission would have summarized this evidence in the Concept Release.

-[7]- See SR-NASD-97-56, submitted to the SEC on July 29, 1997. If approved by the SEC, OATS would impose obligations on NASD members to report to NASD Regulation information regarding orders received by members related to Nasdaq securities. This information would be integrated with quotation information and trade report data in those securities to provide NASD Regulation with a substantially enhanced body of information with which to conduct surveillance for violations of SEC and NASD Rules.

-[8]- Concept Release at 74. For purposes of this comment letter we assume that the Commissionís proposed definition is within the SECís authority.

-[9]- See letter from Frank J. Wilson, Executive Vice President and General Counsel, NASD, to Jonathan G. Katz, Secretary, SEC, dated August 2, 1989, re SEC Release No. 34-26708; File No. S7-13-89 (NASD comment letter on proposed Rule 15c2-10).

-[10]- Delta Release, 55 FR 1895.

-[11]- See Concept Release at 76 -78.

-[12]- In fact, it would appear to us that by moving ATSs out from SRO oversight, the only regulatory authority that will have responsibility for scrutinizing the operations of these "exchanges" will be the Commission. As noted elsewhere in our letter, we believe that a plethora of these exchanges could develop, and the burden of supervising these entities and the novel issues that they will create will fall on Commission staff.

-[13]- While the Intermarket Surveillance Group is helpful in coordinating intermarket surveillance efforts, it only effectively operates after a potential problem has been identified. The creation of additional exchanges would complicate the efforts to identify the initial problem.

-[14]- In this regard, Section 15(b)(8) of the Exchange Act provides that a registered broker-dealer must be a member of a national securities association (e.g., the NASD) unless it effects transactions in securities solely on a national securities exchange of which it is a member. If a broker-dealer that is a member of an ATS non-traditional exchange limits its activity to that "exchange," it would not appear that it would have to join the NASD and be subject to its comprehensive regulatory programs.

-[15]- Although the SEC has attempted to address this concern in the context of the AZX by requiring disclosure by AZX that it is an exempt exchange and is not an SRO, such disclosure does not appear particularly well designed to address the fundamental competitive equity issue and may do more to confuse than inform investors.

-[16]- We believe that the preferable way to narrow the gap between regulatory burdens imposed on SROs currently and those imposed on ATSs is to alleviate the burdens on SROs that no longer serve any necessary regulatory purpose. In this connection, we support the Commissionís suggestion that the SRO rule approval process should be accelerated. Specifically, to follow up on the Commissionís suggestion at page 108 of the Concept Release, we suggest that the Commission consider that new, "pilot program" SRO services and products generally should be effective upon filing, if the new service or product were fully vetted by the SROís governing board that is balanced and represents diverse interests, including both securities industry and non-industry interests. In addition, we propose that such filing also would be subject to the Commissionís right to amend or abrogate the new product or service, after due process is afforded, should the product or service prove problematic under the federal securities laws and rules. In the Concept Release, the Commission suggested that the pilot program approach be limited to "innovative trading programs, apart from their other operations." We believe that this limitation is too restrictive. An SRO may wish to develop an innovative approach to its current mechanism for trading or respond to another exchangeís or ATSís new mechanism. We believe that the SRO should be able to respond rapidly to innovation in the market. Thus, we believe that the SECís rule should be amended to permit effectiveness on filing for new products and services whether they are part of the primary function of the market or in a new area.