October 3, 1997

Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, D.C. 20549

Attention: Jonathan G. Katz, Secretary

Re: File No. S7-16-97

Dear Mr. Katz:

Instinet is pleased to have the opportunity to comment on the SEC's Concept Release, Regulation of Exchanges and Oversight of Markets (Securities Exchange Act Release No. 38672, May 23, 1997).

We commend the Commission on this recent initiative. The Concept Release is another step in the process of continuing to ensure the viability and integrity of the US markets and is also indicative of the Commission's commitment to investor protection.

Previous Releases addressed specific issues in the marketplace. This new Release has greater breadth and depth, both conceptually and structurally -- concerning itself with the regulatory environment of the US markets as well as with cross-border listings and transactions issues. The Commission notes it is “reevaluating its regulation of the markets, particularly its oversight of alternative trading systems, registered exchanges, and foreign market activities in the United States.” [1]

Instinet agrees that the markets are at a crossroads and that it is important to address whether the current regulatory regime, in place for some 60 years, has kept pace with the recent developments in the markets. We also support the Commission's approach for examining these critical issues: a broad-based discussion by all market participants to ensure that as the industry evolves, it does so in the interest of all investors, each of whom relies on the stability of the markets.

Instinet welcomes the opportunity to engage in a dialogue with the Commission and the industry on this important and timely topic. Not only do we agree with the Commission on the current state of the markets, we also agree with the Commission's goal in issuing the Release: it is better to analyze and prepare for the future than to be overtaken by it.

In soliciting comments, the Commission offers suggestions intended to “develop a forward-looking and enduring approach that will permit diverse markets to evolve and compete, while preserving market-wide transparency, fairness, and integrity.” [2] Instinet concurs with the Commission's view that a regulatory structure needs to be in place that is simple, flexible and sufficient to accommodate future business and technological developments. And, with that in mind, we would suggest the most relevant approach to changes in regulation will continue to be based around the fundamental business practices carried out by market participants.

The markets of the United States are the envy of other markets throughout the world. This is owing to the fact that the American markets are at once the most sophisticated and free to be found anywhere. That the United States was able to develop in this way is largely because the regulatory regime that the SEC has articulated over time has preserved the fragile balance between competition and appropriate regulation. Precisely because of the enormity of the task the Commission has set in the Concept Release, thorough analysis of all of the possible alternatives is required prior to the implementation of any structural change.

Instinet believes that the Commission is committed to maintaining a tradition of unfettered innovation, where regulation serves to aid investors and other market participants by giving them the choices they need to improve their performance.

Executive Summary

The SEC has initiated a constructive conversation on issues that are critical to the future of the US markets. Given the current state of the markets and the ideas promulgated in the Concept Release, Instinet believes that there are several principles that ought drive and inform the dialogue moving forward.


  • The US markets have evolved into the freest and strongest worldwide because the SEC has maintained the balance between appropriate regulation and market/business competition. As in the past, an environment where innovation can flourish needs to be encouraged.

  • Although technology is an important factor in determining how markets might grow, a thorough examination of the underlying business practices which account for the behavior of market participants would be most relevant to the discussion at hand. Such an examination would include all market participants responsible for the intermediation of institutional and retail order flow. Regulation, when needed, ought be limited to business practices which threaten fair, liquid, and orderly markets. Obviously, technology does have a role to play in this wider analysis but it should not become the central theme.

Market Practices

  • Healthy competition can only exist between market participants in the same basic business if the playing field is level and fair. As a result, the regulatory regime of the United States ought be structured to ensure that the defining variable is the business activity that a participant undertakes in the marketplace. In deciding on pre-trade transparency/order-handling, for example, any rule or regulation applying to one entity in possession of an institutional client's order ought apply equally to all entities in possession of institutional client orders.

  • Institutions, like individual investors, are entitled to maintain unencumbered property rights to their orders and be able to choose how much and when an order is released to the “market” or “quote.” Consummated trades ought be immediately reported to the broad market.

  • While some trading firms have elected to focus more on proprietary business and while others have chosen agency, both are solid, profitable strategies that use similar technologies, and both deliver services which investors value.

  • Efficient markets need to strike the appropriate balance between “upstairs” trading (with immediate trade publication to the relevant exchange) and “downstairs” trading. Clients are demanding the flexibility of using both of these mechanisms when executing their global securities. National markets that do not strike (by mandating all orders onto the “floor” or into the “quote”) this appropriate balance will lose market share to those national markets that allow such flexibility.

  • Continued quality of markets is contingent upon a variety of issues beyond technology. The issues of market manipulation, fraud, surveillance and capacity relate in large part to issues that apply to trade processing in all firms. Consequently, this issue calls for an examination of how trades are processed throughout the industry.


  • Due to the increasing convergence of the main market participants, Instinet does not believe that anyone will be able to continuously or consistently draw the lines between entities in response to new technologies or market developments without causing an unlevel competitive playing field in the process.

  • Specifically addressing the model proposed in the Concept Release, Instinet believes the Commission should widen the scope of its analysis. We believe that a simple structure which redefines the traditional broker/exchange paradigm and reduces the proposed “tiers” is a model which merits consideration. This new structure would be organized around two “boxes” – an “SRO member/exchange member” and an “investment exchange.” Incorporating aspects of new international regulatory models, and allowing for the reality of the “blurring” of the lines between market participants, the key element in Instinet's alternative proposal is choice – true choice.

  • Rather than attempt to firmly define which “box” a firm should operate in through a complex and subjective analysis based on technology, dissemination, type of order or function, Instinet suggests that the entity itself choose its regulatory “box,” based on its positioning in the marketplace and its service offerings to clients. Each “box” in our alternative model would have specific rules and regulations giving firms the option to decide which “box” is best suited to their business. The overall result, we believe, will be increased competition and innovation, and continued growth in the US markets.

  • Concerning foreign share trading, Instinet believes that the Commission should investigate the wider industry practices as they relate to this type of activity and not limit its focus on technology. Instinet also believes that non-US order flow is appropriately mediated through the mechanism of a regulated US broker. In regard to any additional regulation that ought to be put in place when a non-US regulated broker is involved, we would suggest that the Commission consider designating markets as relatively “safe” or “less safe” for US investors.

  • Regarding individual foreign security issues, Instinet believes there is too much regulation being placed on market participants trading in inherently less risky shares (GDRs listed in US dollars on the London Stock Exchange) and perhaps not enough “investor protection” when these same entities are trading local shares directly in the emerging markets.

In sum, using business practice as the organizing principle for an analysis of the current market and as the basis for determining the regulatory needs for the market in the future will, in our opinion, achieve the Commission's goals. Developing a simple and flexible model for the future, backed by rigorous regulatory scrutiny and enforcement, will incorporate the best of the current market and ensure continued innovation for the future. These principles and ideas are expanded upon below.

Instinet and the Global Competitive Environment

Throughout its 28 years Instinet has consistently supported local stock exchanges. During its history, Instinet has evolved so that regardless of market structure or regulatory regime, the firm's provision of agency trading services for its clients has proven to be successful. In some markets Instinet, like our competitors, applies technology to this basic trading service. In many markets, however, the firm trades simply on a telephonic basis. To characterize Instinet as an automated trading system (ATS) misses the true basis for the firm's success. That success is wed not to technology, but to the reduction of trading costs (and the attendant increase in investment performance) for its clients. Instinet trades in 40 markets around the world -- often in markets where trading is not automated. Never can our business or our firm be reduced to being a mere system.

Instinet's mission is no different from any other global institutional trading firm: to secure as many institutional orders as possible and execute them at the best prices possible anywhere in the world. In order to fulfill this mission Instinet, like all other firms in the equity trading business, uses its human and technological trading skills to enable its clients to secure best execution. All firms that manage institutional orders during market hours use the same basic fundamental strategies for executing a large trade. That strategy is to represent pieces of the client order simultaneously in as many different pools of global liquidity as possible.

In many equity markets Instinet, like many of its competitors, represents pieces of client orders in multiple markets for that security simultaneously, thereby giving clients access to exchange member liquidity. For example, if a firm is handling a client order in a French share such as Elf Aquitaine, pieces of this order would be represented on CAC, SEAQ International and the New York Stock Exchange simultaneously.

While sending pieces of client orders down to the relevant exchanges all firms (Instinet included) are advertising pieces of these orders to the world's fund managers and block desks, thereby tapping into these important sources of “upstairs” liquidity. Using technologically-based facilities such as the FIX protocol or telephonic means, many of these firms disseminate firm bids and offers out to a wide group of global clients. If a client then wants to hit a bid or take an offer, this can be done by either hitting a few key strokes or responding by voice.

Large integrated trading houses also, by using their own capital, facilitate client trades through advertising and trading blocks out of their own “book.” For example, if a large firm is “long” 200,000 shares of IBM, it will either electronically or telephonically send messages to its global client base that it wishes to sell 200,000 shares of IBM at a specified price. After an “upstairs” trade occurs (whether it involves dealer capital or is a pure agency cross) it is then reported to the relevant exchange or SRO in accordance with its trade and interaction rules. Currently, a substantial amount of the daily volume of the major global exchanges (including the NYSE) is traded in such a manner. Increasingly, institutional clients are demanding that their brokers use electronic means (rather than telephonic) to send them these firm “upstairs” bids and offers.

What separates institutional trading firms from one another is not what they do, but rather how well they implement and execute these strategies on behalf of clients. For instance, firms with a strong local presence in foreign markets typically offer a more competitive service to their clients than those who do not. Moreover, firms that have a large global institutional client base tend to trade more blocks “upstairs” for their clients, enabling them to tap into this important source of liquidity.

Instinet separates itself from competitors in two ways: first, and most importantly, by being a pure agent; second, by applying advanced technology to the trading process when and where it can to increase overall trading and operational efficiency. It should be noted that in many markets around the world, the application of technology for this purpose is not possible for a wide range of factors (mainly the protection of the local broker community). Thus, in those markets Instinet offers a pure telephonic agent service.

It is important to understand that Instinet made a business decision to be solely an agent just as many of our competitors, who had historically led with their agency business, recently decided to lead strongly with proprietary trading. Starting in the late 1980's many of the “bulge” bracket firms found their traditional agency business facing shrinking margins as commission rates were reduced and clients were demanding more control over the trading process. These entities, therefore, decided to put more of their firms' financial and intellectual capital towards the higher margin proprietary trading business while simultaneously upgrading their traditional research, market making and investment banking capabilities. This strategy has been very successful for many of the large integrated brokerage houses.

Instinet, on the other hand, saw an opportunity to apply advanced technology to what was perceived as the outmoded and increasingly unprofitable agency business. Through this combination of diminishing competition for agency business, and the efficiencies of technology that Instinet employed, this too has been a very successful global strategy. At this point in time Instinet is the only pure global agency broker. However, on a daily basis Instinet faces keen competition from hundreds of integrated houses for institutional order flow. Instinet's business positioning is what makes us unique and not systems, processes or technology.

The Concept Release

For its part, the Commission's view is that the changes that have occurred in markets are owing to both the evolution of business practice and the technologies now employed across the board both domestically and internationally. Legitimate and important questions have been raised regarding the regulatory framework for, and oversight of, the current technologies and the impact that these technologies have on the investing public.

Of particular concern to the Commission is the technologies that drive so-called alternative trading systems. Currently these systems are for the most part regulated as broker-dealers. It is under this paradigm that the Commission wishes to examine changes to the present regulatory model.

To remedy the perceived deficiencies in the current structure and to provide for quality markets and investor protection, the Commission proposes two models for the operation of these systems.

The first model is a tiered structure in which ATSs may function as for-profit exchanges and assume SRO functions. The placement of an ATS in a particular tier would be driven by the volume executed by the ATS, the dissemination of certain types of orders and whether the ATS contains a certain type of price discovery mechanism.

The first tier would represent small volume providers and be exempt from full exchange regulation. ATSs that have so-called passive price mechanisms would also be included in this tier.

The second tier would represent large volume ATSs that permit price discovery. In addition, a distinction will be made as to the types of orders (priced/unpriced) and how widely these orders are disseminated. What appropriate SRO functions these systems would assume, given their for-profit positioning and their private sector ownership, is open for comment.

The third tier is reserved for traditional exchanges. The questions raised here regard the possibility of eliminating regulations that are preventing these exchanges from innovating or competing more effectively.

The other model proposed by the Commission is the continued regulation of ATSs as broker-dealers with enhanced regulatory oversight and supervision by existing SROs. Under this model, the SEC would propose regulation that in the Commission's view improves transparency, surveillance and capacity of these systems.

Finally, the Concept Release attempts to deal with the issue of investor protection when trading cross-border. This section focuses on the access points for entry of foreign trading into the US, and asks questions regarding the types of reporting, record keeping, and disclosure that might be necessary to achieve the Commission's goals.

These ideas are addressed below.

Business Practice and the Concept Release

The natural and inevitable evolution of firms applying technology to their niche(s) of the business has been very healthy for the US market and all of its participants. However, we believe it is essential to understand that technology is implemented to reflect common business practice but does not fundamentally change business practice.

Because of its comprehensive industry-wide nature, Instinet takes the position that business practice should form the basis for regulation rather than technology or volume. Both currently and in the future, this will enable the Commission to apply the same standard to all market participants engaged in the same fundamental business. It will also allow for a simplified and flexible regulatory regime and market model for the future.

Instinet is concerned not only with the impact of underlying business practices as a critical issue for examination at this time, but also with the real possibility of one set of firms in the same basic business being placed by (present and) future regulation under a competitive, client service, and flexibility burden which others in the same business will not share.

In addition, the effect of any existing or new regulation should not reward industry practices which do not advantage institutional and retail investors. For example, the practice of payment for order flow, which is the centerpiece of existing US retail practice, deserves serious investigation as it gets to the heart of whether retail investors are in fact receiving “best execution” on their order flow. We would also like to suggest that the Commission put under close scrutiny the questionable and active trading that frequently takes place in the days and weeks before a large deal is announced. This type of activity seems to be becoming quite commonplace and dealing with this issue gets to the heart of a market's integrity. Instinet would urge the Commission to continue to thoroughly investigate these questionable industry practices.

To make their businesses more efficient, increasingly technology is being used in and to communicate amongst firms. Bulge bracket trading firms are using innovative technology to aid them in their research and IPO distribution capabilities. Brokers are now able to send their individual research reports to a filtered yet widely disseminated group of global clients. In addition, many of these firms are also using technology to increase the efficiencies of their “book building” process when bringing a new company to the market.

Practically speaking, there is no reason why a firm that uses capital could not roll out more of its internal trading technology (and many of these firms are in the process of doing so) to all clients so that “upstairs” and exchange routing interactions are made more accessible and visible to their clients. It is essential to understand that the rolling out of technology is ultimately a business decision. Instinet chooses to make more of our technology available to clients to bring further efficiencies and control to the trading process.

If the appropriate balance between business practice and technology is not achieved when dealing with new regulation, then there is significant risk of providing advantages to some market participants over others who are engaged in the same business practice. This alone will seriously harm the Commission's goals and programs.

Unfortunately, we are already witnessing the beginnings of such a situation with the implementation of the order-handling rules. Market makers, ECNs, agency brokers, periodic crossing systems, as well as new entities such as OptiMark are (and will be) in intense competition to win business from institutional investors. A level regulatory regime must offer all of these firms the same menu of choices in terms of order-handling/display when executing institutional orders.

At present, market makers, ECNs and agency brokers have limited choice as to the display of certain client orders “on exchange” or “in the quote,” whereas other entities, such as so-called traditional brokers and periodic crossing systems (OptiMark included), do not have any obligation to show any part of a client order “on exchange” or “in the quote.” All institutional trading firms must have the same choices in offering clients a flexible and functional execution service. Given similar choices, each of these entities can then construct their own menu of tailored solutions in order to win client business. Healthy and level competition amongst these entities should determine who wins and loses business; not the regulatory regime.

Stakeholders and the Blurring of the Lines

The lines between the three traditional stakeholders in the market: investors (particularly institutional investors), brokers and exchanges have become more porous over time. Large institutions -- through internal crossing systems and ownership of brokerage firms -- are beginning to exercise brokerage and even exchange functions. Conversely, large brokers -- through “upstairs” block trading and internalization of retail order flow using dealer capital -- are emulating the traditional exchange trading function and -- through the acquisition of asset management businesses -- are also emulating their large institutional clients. Finally, exchanges -- via their sophisticated order management systems and “sponsored” terminals -- are emulating the traditional brokerage function.

This natural evolution and blurring of the traditional lines of demarcation have been healthy for the overall market, and has lead to an explosion of liquidity and innovation in the US. While we believe that it is time to investigate a new regulatory framework and we applaud the Commission's efforts, Instinet does not believe that the Commission or anyone else will be able to draw lines between the various types of mechanisms currently used by investors to execute US and non-US order flow. Further, we believe that any attempt to draw absolute lines between in essence Siamese twins will, at best, disadvantage some entities over others, and at worst, create an infinite amount of regulatory categories for firms engaged in the same basic business.

The Commission and Instinet seem to agree that trying to define the absolute difference between a “broker” and an “exchange” is becoming increasingly difficult, and will only get more so in the future -- and, we would argue, impossible to get right.

Convergence and a “Tiered” Approach

Instinet is concerned that the current “tiered” entity proposal is overly complicated and, as such, will create undue regulatory and (more importantly) client flexibility burdens on entities in the same basic business. This will lead to a cacophony of regulatory regimes without a strong central theme. Each of these tiers will need to be designed to delineate the role and nature of a particular technology. Technology is a necessary element in the regulatory mix, but is insufficient to serve as a central organizing principle for regulation. This burden ultimately will impact the quality of markets, effecting all market participants, including regulators.

There is a distinct possibility that if each new technological innovation faces large regulatory hurdles, both existing market participants and innovators will, in fact, be disincented from developing such technologies, and the markets will suffer. In addition, any potential alliance between entities that could produce efficiencies for the markets, either domestically or internationally, will be forced to consider a potential regulatory backlash. Further, regulators will have to develop new regulatory regimes for technologies or other market developments that do not fit into the proposed model. The result will be many tiers, exempted entities, traditional entities, and as yet unknown constructions beyond what are now known as brokers and exchanges. How many exchanges and SROs can the SEC realistically permit and oversee?

There are other issues specifically regarding the proposed tiered structure that relate to level competition, transparency and innovation. Under the present proposal, firms that represent institutional investors are being sent a strong message that the less an entity contributes to the price discovery process (pre-trade transparency) the less regulation it will receive. For example, it is our understanding that under the proposed regime, firms such as POSIT and OptiMark will not have the same order-handling/display requirements as traditional firms or so-called ECNs. POSIT and OptiMark are in the same basic business as any other firm executing client orders. To have separate transparency regimes for these firms will result in an unlevel competitive playing field.

In addition, for the reasons stated above, we question whether any new regulatory regime should allow tier one and two entities to have a lower threshold of regulatory, and more importantly, order-handling/display functionality than entities in tier three. This is particularly true since firms in tiers' one, two, and three are fundamentally competing for the same business. Finally, as tiers' two and three will be divided by volume, practitioners will be incented to enter into regulatory arbitrage opportunities at the “cusps” or margins of the volume/price dissemination cutoff between the tiers.

Globally, our other major concern with the proposed regulatory model is the effect being placed in a different (non-broker) “box” will have on an entity's ability to continue to operate “normally” in overseas markets. The US markets are the most advanced, competitive and complex in the world, therefore looking at a new regulatory model at this point in our evolution makes eminent sense. However, in most markets around the world -- particularly in emerging markets -- the present regulatory model (especially the blurring of exchange and broker) is not under any stress.

For example, the majority of markets in Asia have one national exchange by law (frequently it is floor based) and ”traditional” brokers. If any institutional firm is labeled anything other than a “broker” in the US, how will this effect a firm's ability to operate in markets that do not have such an evolved regulatory regime? This is an absolutely critical issue for the Commission and all US market participants to deal with as we continue to refine our ideas on a new US regulatory framework.

We are also concerned with any regulatory regime that places too much emphasis on exemptive authority. Regulation should, in our opinion, offer broad sweeping yet simple rules with few exceptions, and not be based on offering exemptions from complicated rules. To implement the “line-drawing” model proposed in the Concept Release will impose undue regulatory costs and burdens on all entities (especially the Commission), and unintentionally create an unlevel playing field amongst entities in the same fundamental business. These questions argue again for similar requirements for all trading firms and one standard across the board.

Regarding Transparency

In its Concept Release, the Commission argues for greater integration of so-called ATSs into the NMS, particularly as ATS trading facilities are executing substantial amounts of order flow and as information about that order flow is not available to the broad market.

While we believe these to be legitimate issues, we also believe there to be a more fundamental issue: what information regarding “upstairs” order flow, regardless of venue, is material to the market as a whole, and when must it be released? The answer to this question would then incorporate into this discussion not only ATSs, but also every other entity in the institutional and client execution business.

In the past two decades the amount of money intermediated throughout global markets each trading day on behalf of investors has risen dramatically and technologies have been applied at all points in the trading/processing cycle. As discussed earlier, a substantial portion of investment is intermediated by trading firms through negotiation and trading of large blocks of stocks “upstairs” which are then reported to exchange floors. Virtually all trading throughout the world is now subject to some technological “handling”-- brokers' technology that internalizes and executes order flow, passive matching services, order routers, and information vendors. Each offers a means that investors routinely use to garner information and secure best execution.

There is no difference in either degree or kind concerning transparency issues in these various trading mechanisms. So, directly related to the question of investor protection in the current environment, is the relative transparency that each of these mechanisms provides to all investors in the market.

That being said, as a practical matter, exchanges have found that the mechanism of “upstairs” trading in fact contributes to overall market liquidity and that this block trading leads to greater market efficiency as well as increased client flexibility, access and price discovery. All market participants benefit from a consistent “upstairs” trading regime as this type of activity, if coupled with immediate trade publication, significantly increases overall liquidity in the markets.

Paradoxical though it may be, “upstairs” trading makes markets more fair, liquid and orderly than in markets where it is not available. Mandated pre-trade transparency (forcing members to send the entire client order down to the “floor” or the “quote”) results in illiquid markets, as this stifles investor flexibility in seeking to protect the legitimate property rights of their orders. These rights include all the pre-trade research and information that defines a trading strategy, as well as the decision on how much of an order to show and to whom at any point in time.

A critical component of any effective trading strategy is the ability to decide when and how client orders are released to the broader market. The Paris Bourse found that when it mandated that all of its members send all client orders “to the floor” to be matched, thereby eliminating anyone's ability to trade “upstairs,” liquidity decreased and fled the market for other venues (primarily SEAQ International), which offered this flexibility. When Paris reintroduced an “upstairs” trading regime with immediate post-trade publication to the exchange, liquidity was repatriated. The Japanese market is, at present, experiencing similar market structure problems due to its insistence that all trades executed by TSE members need to be physically “matched” on the exchange.

A significant percentage of daily turnover in Japanese equities is being executed by member firms in London rather than Tokyo simply because SEAQ International allows and encourages “upstairs” trading. Clients and exchange members “vote with their feet.” They both operate in hyper-competitive businesses and must be ruthless about getting best execution on their order flow.

The simple fact of the matter is clients need the flexibility to trade both “upstairs” and “downstairs” and efficient market structure (and any regulatory regime) must ensure that this is offered in a flexible yet orderly manner. Institutions that are managing increasingly large pools of assets need the flexibility of both “upstairs” and “downstairs” trading in order to efficiently buy and sell large blocks of securities. Those national markets that offer this flexibility will be the natural winners in bringing member and client liquidity to their markets. However, those national markets that stifle this flexibility will, over time, lose market share to other national markets.

Moreover, all members of exchanges that offer “upstairs” trading should operate under the same rules regarding their ability to piece out client orders, and their choices for when and how much of a client order to show “on the floor” or “in the quote.” Any regulatory regime that does not permit those in the same business to operate under the same rules regarding order-handling/display will be fundamentally flawed, and will naturally preference those entities that are able to offer more choice/flexibility to their clients.

The Commission's Release settles on the need to adjust the regulatory status of so-called ATSs owing in part, to their technological base. Yet ATSs merely utilize technology to reflect common industry practice. It is entirely possible that by focusing the regulatory telephoto lens on technology, while not stepping back and refocusing on the wider industry practice, the real issue will not be properly addressed.

Instinet, therefore, would encourage a broader based discussion of the overall issue of transparency in “upstairs” trading. Moreover, Instinet would argue that in order for markets to be fair (and not advantage one set of firms over another), that any transparency regime apply equally to all. In order for there to be a consistent market-wide regime, dissemination of pre-trade order information and order handling must be applied equally to all market participants conducting “upstairs” business. A rule that would apply to any market participant in possession of order information would logically apply to all market participants holding order information.

The market demands consistency from the Commission on this issue:

  • either achieve pre-trade transparency by mandating that all orders from all firms handling institutional and retail business must be “on the floor” or “in the quote”;
  • or, the preferred option, that consistency be achieved by mandating immediate post-trade publication of “upstairs” executions to the relevant exchange -- leaving the choice as to what is shown “on exchange” by a member up to the individual entity and/or client handling the order.

For the reasons stated above, mandating pre-trade transparency across the board will have the result of negatively impacting overall liquidity in the market and will drive trading in US shares to other national markets that will offer more flexibility in this area.

The post-trade transparency approach, which gives the end-client and the exchange member choice of how much of an order to display “upstairs” or “downstairs” as well as when to send this order down to the relevant exchange, if combined with immediate post-trade publication transparency, seems to be the model to which most efficient exchanges are gravitating. It is then up to each individual exchange to build an appropriate mechanism that incents but does not mandate (through monopolistic rules) that members and their clients send orders down to the market for execution.

Markets such as Stockholm, Hong Kong, and Paris (amongst others) have over time, via a consistent post-trade transparency regime, achieved the appropriate and delicate balance between “upstairs” and “downstairs” trading with the result being truly efficient markets for both clients and members.

Striking the appropriate balance on this issue is essential if the US markets are to remain the most liquid in the world. Instinet would welcome an industry-wide discussion of this critically important issue.

Globalization and Cross-border Trading

The Concept Release raises a number of interesting issues regarding the trading of non-US equities by US-based clients. While it is essential that US investors are “protected” when dealing in foreign markets, it is also critical that these US investors are able to apply increasingly sophisticated routing and trading strategies to their international trading process.

US pension fund holdings in foreign shares are expected to grow substantially over the next few years from approximately $475 billion in 1996 to $940 billion by 2001. [3] It is absolutely essential that, for the continued competitiveness of the US-based international fund management industry, these managers are able to employ the most cost effective techniques when executing their foreign trades.

In one fundamental way, the global situation parallels that in the US: that is, it is the underlying business practices in the securities industry that ought be the focus for examination and appropriate regulation, rather than the technical delivery mechanisms that the various business entities employ in implementing their business decisions. The fact is that there are a myriad of issues regarding order flow in foreign equity trading and that systemic industry-wide practices should underlie the concern for investors and regulators rather than any particular investment instrument or technology.

There is a degree of irony in the present level of regulatory “sufficiency” in some situations. For example, if an order in a German share is given by a US-based fund manager telephonically to a US registered brokerage in New York, which then sends this order to its London office, which then forwards the order to its Frankfurt office (or to a sub-agent, if the US broker is not a local exchange member), where the order is executed on IBIS (the electronic component of the Deutche Bourse) -- that trade is deemed appropriately regulated. However, if the same US-based broker places a sponsored IBIS machine (similar to a sponsored DOT machine) on the fund manager's desk, allowing the manager to electronically send the order “directly” down to IBIS for execution, additional regulation is required.

The sponsored IBIS scenario offers the client better access to the market and more information from the market. In fact, such a scenario poses fewer problems (both from a regulatory and business perspective) than does the telephonic practice. With a broker-sponsored IBIS terminal, investors exercise measurably more control over their orders, there is increased transparency, greater liquidity, plus an electronic audit trail - all of which are goals the Commission seeks in terms of investor protection and market quality. How can one suggest that additional forms of regulation be applied to a cleaner, more transparent process?

There are, however, legitimate oversight issues when a US-based broker is not involved in this process. Instinet believes that in the interest of investor protection there should be an appropriate, but not onerous, “vetting” process for non-US registered brokers to offer their services - whether it be telephonic or electronic (sponsored IBIS terminal) -- to US fund managers trading foreign shares. We would suggest that if the foreign broker is already a member of a major international stock exchange (the London Stock Exchange for example) and wishes to offer US-based clients a UK trading service -- then no additional regulation should be required. However, if a foreign member of an emerging market wishes to offer US-based clients a local trading service then additional safeguards or “warning labels” may be necessary.

Instinet supports the view that the regulatory standards to which US registered brokers must adhere are appropriate, and that US investors are entitled to the assurance that their contracting broker is “vetted” whether they do business over the telephone or through usage of broker-sponsored exchange terminals. The oversight issues are not related to usage of technology but rather to the business practice and responsibilities of the relevant local broker.

Next, again focusing on technology rather than business practice, there is a view that front-end integrators of broker-sponsored exchange terminals need regulation. If an information vendor, for example, wants to integrate multiple IBIS gateways at the client (fund manager's) front-end, we do not see a need for any regulation. The fund manager is still contracting with the executing brokers that are chosen and the brokers are still responsible for following all the rules and regulations of the local exchange. If one were to regulate the systems' integrator it would be roughly analogous to regulating the manufacturers of television sets rather than the television networks. These integrators require no regulation, as the major service provided is the ability to reduce desktop real-estate and provide for efficient order transmission to a broker.

The final, and perhaps most interesting question regarding intermediation is raised when a foreign stock exchange wishes to place its own terminal on a US-based fund manager's desk and allow that fund manager to send orders, without any broker intermediation, directly to the exchange.

Instinet feels that as long as the appropriate investor protection, compliance, capital and regulatory standards have been met, then this type of entity should be able to offer its services to US-based clients as long as it becomes a US registered investment exchange. The Commission should ensure that a US-based fund manager is at no less risk when executing trades directly via an exchange terminal than when trading through a broker sponsored exchange terminal.

In this vein, perhaps the regulator could, much like the State Department, evaluate the relative safety of foreign markets on behalf of US investors. There would be markets that are “safe” to trade in, questionable markets, and markets that investors ought avoid. The Commission would not stop investment, simply issue appropriate warnings to investors.

Regarding the issue of GDRs, we again find it ironic that if a US-based fund manager wants to trade, for example, an Indian ordinary share through a registered broker then no regulation or “warning label” is required, despite the inherent settlement and currency risks encountered in any emerging market. However, when the same US-based fund manager wants to trade an Indian GDR, which is traded and settled in US dollars and listed on a “safe” exchange such as the London Stock Exchange, substantially onerous regulation is required. This is illogical as GDRs traded on the London Stock Exchange are no more or less risky than any other security listed on that market, and they are substantially less risky than emerging market ordinary shares.

Consequently, Instinet believes that the options outlined for GDRs do not achieve the Commission's objectives, and a more in-depth analysis is warranted.

Systemic Risk and Surveillance

Given the breadth of communication between market participants and the increasing volume growth in the US markets, limiting systemic risk becomes an increasingly difficult and important issue. For this reason, Instinet believes that the “capacity” debate should not strictly be centered around technology nor around the premise that competitive forces will be inadequate to ensure participants maintain adequate operational capacity. For these reasons, Instinet would urge the Commission to take a broader more industry-wide approach to addressing these very important issues.

Systemic risk can arise from any or all market participants who are involved in trade execution and processing. Simply because one participant may be responsible for a significant share of activity in any one market does not alone constitute operational “risk.” In fact, participants that operate -- whether traditionally or by applying technology to their business - in any one particular niche of the market could also be considered potential contributors to systemic risk if they are suddenly unable to perform their specialty service. Firms that have significant market share in any one particular issue would fall under this category.

Competitive forces have created multiple avenues in which participants can perform trade processing. For example, market participants can choose to use the telephone, order routing systems, proprietary blotters, NYSE's DOT, and SelectNet (among others) to reach various pools of liquidity within the market to execute trades. If there is a “shock” to any one liquidity pool, other avenues may be used to maintain the required stability participants have come to expect.

Over time, any market participant unable to provide reliable, fast, efficient and error-free trade processing will be replaced by competitors that can. Forcing potentially rigorous regulation as to what loads of trade processing participants must be able to handle will at best create more complexity and at worst create a less flexible competitive landscape with more onerous barriers to entry.

The issue of more stringent surveillance of so-called ATSs was also raised in the Concept Release. In particular, the Commission believes that there are deficiencies because “the SRO does not receive a composite picture of orders available on that alternative trading system on a real-time basis. Consequently, the SRO is not able to integrate the activity on an alternative trading system into its information about activity in that security on its own market.” [4] Furthermore, the Commission believes that without this level of information, there is greater risk of fraud and market manipulation. Instinet disagrees with these contentions.

First, there is absolutely no evidence of any fraud or manipulation on the part of so-called ATSs or ECNs of which we are aware. In fact, the visibility and audit trail within firms that apply technology to the trading process is significantly greater than in firms who choose to operate in a less advanced fashion.

Second, the types of information Instinet makes available to its SRO are no less comprehensive than that of any other SRO member. In fact, as detailed above, Instinet would argue that the content of the information it provides is indeed greater. Therefore, Instinet believes it should not be treated differently merely because its technology has afforded better audit trails, and that any requests for this kind of information be required from all participants that are in the same fundamental business.

In conclusion on this topic, concentrating the Commission's efforts on firms that apply technology rather than on firms who do not are, in our opinion, backwards.

A Variation on the Commission's “Model”

Instinet applauds the Commission for initiating a discussion on important regulatory issues. The Concept Release raises significant questions which require thoughtful responses from all market participants. We also share the Commission's sense of the timeliness in reviewing the US regulatory posture. In addition, we strongly believe that prior to implementing any structural changes, consideration of all alternatives is imperative.

The fundamental role of any regulatory regime is to ensure that the markets are properly regulated. The simplest and broadest approach, articulated in 1934, based the regulatory/business practice regime around two well defined entities - brokers and exchanges - each expected to fulfill different and important functions in the marketplace. In the main, exchanges were designed to provide a locus for a fair, liquid and orderly marketplace, as well as to list and vouch for the quality of securities listed. Brokers were assigned the responsibility of representing investors in the marketplaces. This model has served the markets well for over sixty years.

However, as previously noted, the “lines” between these entities have become more porous over time. The challenge that we now face is to decide whether a regulatory regime still needs to be centered around the drawing of lines for each functional entity, or might we be better served by developing a simple and flexible model reflecting the reality of “blurring lines”. Instinet believes that it is time to construct a new regulatory model based on the latter.

It is our view that a regulatory regime conceived in this fashion will most benefit the US markets. This slightly different approach redefines and expands the parameters of the current broker/exchange model. It is our opinion that the labels - “broker” and “exchange” - have become obsolete in today's world. These terms should be replaced by “SRO member/exchange member” and “investment exchange” to reflect their broader and potentially more interchangeable roles.

There is a critical yet subtle difference between this approach and both the 1934 model and the “tiers” articulated in the Concept Release. That difference is rather than attempt to define which “box” a firm should be in through a complex and subjective analysis based on technology, dissemination, volume, order type or function, instead each entity would choose the “box” in which to operate based on the service it offers and its competitive positioning in the marketplace.

The Commission would then establish appropriate operational guidelines for “SRO member(s)/exchange member(s)” and “investment exchange(s).” It would also ensure that inter-exchange linkage and transparency regulations were in place to take into account the addition of new “investment exchange” entities.

For instance, if a firm chose to become a member of an exchange such as the New York Stock Exchange, then it would have made a fundamental business decision to follow the rules and guidelines of this properly regulated entity (the NYSE investment exchange). However, if the same firm wished to offer a trading service to its clients, but for positioning/marketing reasons did not want to be a member of the exchange, then it should be able to choose to become an investment exchange itself.

All investment exchanges, under this alternative approach, would then be subject to any/all inter-exchange linkage and transparency rules. In fact, the existing market structure, through the Consolidated Quotation System (CQS) and the Intermarket Trading System (ITS), already has an excellent mechanism in place to link all investment exchanges. CQS provides consistent inter-exchange information on pre-trade prices as well as post-trade publication while ITS provides a powerful inter-exchange execution capability.

So, to continue this NYSE scenario, if an existing member of the NYSE chose to become an “investment exchange” trading NYSE shares, another exchange line would appear in the consolidated quote line (CQS) and through ITS the necessary inter-exchange execution linkage would be in place. Similarly, an ITS type of linkage could be developed if multiple investment exchanges chose to trade US OTC shares. Inherent in this paradigm, however, are the thoughts that artificial “listing” competition ceases to exist, and that all listed and OTC securities be freely traded by any investment exchange. It is our view that this would result in true inter-exchange competition and foster innovation in the US marketplace.

The scenario discussed above is beginning to unfold in Europe. Tradepoint, a London-based firm, chose not to become a member of the London Stock Exchange. For positioning and marketing reasons, the firm preferred the designation of a Recognized Investment Exchange (RIE), although it is in the basic business of “winning” institutional order flow. However, Tradepoint could, if it wanted to, join any major European stock exchange tomorrow and be in compliance with exchange rules and regulations without having to change any of its technology or order handling functionality. Conversely, any member of the London Stock Exchange could choose tomorrow (for marketing or positioning reasons) to “pull out” of the LSE and, without changing any of its existing methods of handling client orders could “become” an “investment exchange” overnight.

The critical missing link in the present UK regulatory model is that no inter-exchange infrastructure is in place. The UK regime presently lacks a consolidated tape of all “investment exchange” pre- and post-trade prices as well as an ITS inter-exchange execution facility. Nevertheless, the Tradepoint/LSE member scenario does suggest some interesting points for consideration. First, that there is increasing recognition that there is little if any functional difference between a for-profit exchange that offers direct institutional access, and a member of a large traditional exchange or SRO. Second, why not embrace the concept that the lines between the various market participants are blurring and construct the regulatory framework around this development?

Also, there is a strong global trend towards for-profit, direct institutional “access” exchanges (AZX, Tradepoint) and demutualized (Australia, Stockholm and Milan) exchanges. These are very interesting and innovative developments and they raise a number of important issues relevant to the Concept Release. The most important of which is that these new entities are making the traditional demarcation lines between broker and exchange even more porous, and that these lines will only get more blurred looking forward. In regards to being able to draw firm lines between functional entities, the straw that is breaking the proverbial “camel's back” is this for-profit, direct institutional access exchange which trades, but does not list securities.

We would urge the Commission, and the entire industry, not to attempt to base a regulatory model for the 21st century on attempting to draw lines between entities that have evolved into twins. We must make sure that the markets are properly regulated, but we feel it is time to approach this whole issue from a slightly different angle: give all entities in the marketplace the business choice as to which “box” they want to operate in, while making sure there is an umbrella of proper and consistent regulation tying all markets and participants together.

A strong inter-exchange foundation (CQS and ITS), put in place by the Commission, already embraces a structure offering true competitive choice. Instinet believes that the alternative we are suggesting takes advantage of this strong foundation and offers a simple yet comprehensive regulatory regime.

For this approach to succeed the Commission needs to insure: that an “investment exchange” can not put in place rules or regulations that stifle inter-exchange competition; that the “investment exchange” box be designed as a fair and orderly market; and, that a strong inter-exchange linkage mechanism is in place for pre- and post-trade data and executions.

If there is a strong yet flexible regulatory regime in place for “investment exchanges,” then ensuring that “SRO members/exchange members” are properly regulated becomes relatively simple. If the exchange itself has the appropriate rules, capital requirements and safeguards, then any entity that chooses to follow those rules will, by definition, be properly regulated.

We understand that there are, of course, a number of important issues to discuss within this slightly different approach. However, we believe that by and large the Commission's proposed model by taking this slightly different approach, has the makings of a comprehensive, simple and flexible model that will position US markets for even more rapid innovation than has previously been the case.

Instinet would welcome continued discussions with the Commission on this critically important issue. The one and only parochial interest we do have is that under any new (or existing) regulation, the Commission must make sure that rules do not give a business advantage to one set of firms over another who are in the same fundamental business.


The US regulatory regime of promoting competition and technology is the envy of the world and has led to significant growth in our markets. Tremendous innovations have taken place.

Instinet agrees that it is an appropriate time to analyze a new approach to adapt to these changing times. The ideal environment is one where regulation assures fair, liquid, and orderly markets. At the same time, such an environment needs to foster competition and reward innovation. This is a complex challenge, but one that must be achieved if the US is to retain its enviable global position.

Regardless of structure, good markets are based on integrity, investor protection, appropriate transparency, breadth and depth of liquidity, and ease of round-trip access. The market structure itself must provide for and reflect each of these key elements.

Instinet is of the view that at the present time dialogue is of the essence, so that all the potential implications for changes are adequately aired and examined. Also, this will ensure that broad central organizing core principles for the future of market structure will be agreed to prior to implementation of any specific model beyond current models.

Such core principles must endure beyond any technology, and much like the US Constitution, set guidelines that are not tied to any particular time or place. Therefore, the regulatory regime must be applied to the business practices that are conducted in the marketplace. To assure market fairness, the Commission needs to consider taking a wider approach to new regulation, the central theme of which would be industry-wide business practice as well as fairness and investor protection. Any existing or new regulation must not have the effect of giving a business advantage to one set of firms over others who are in the same fundamental business. This is particularly critical in terms of any (existing or) new order-handling, “upstairs” trading, or transparency regime that is put in place.

A slightly different approach based on the Commission's excellent proposals should be investigated. We would like to suggest that a simple and comprehensive regulatory structure be developed which embraces the reality of “blurred lines” amongst traditional market participants. With such an approach, each business entity could choose the regulatory category appropriate to the service it provides to its clients. The already strong foundation of inter-exchange information and execution linkages would provide the appropriate regulatory umbrella under which healthy and level competition for share trading would occur between “investment exchanges” and amongst “SRO members/exchange members.”

And given the fact that the most effective regulation is simple, strong, flexible and broad - basing regulation on exemption would indicate that the regime is by definition either too complex or not encompassing enough. We agree with the Commission that regulatory principles must assure that investors are protected and markets have complete integrity. In that spirit, all market practices and mechanisms need to be included in the Commission's analysis.

Instinet looks forward to continuing to contribute to this important dialogue.

Sincerely yours,

Douglas M. Atkin


[1] Concept Release (File No. S7-16-97) at 10.

[2] Concept Release at 10.

[3] InterSec Research Corporation - Global Pension Statistics (1997)

[4] Concept Release at 57.