Via e-mail at

May 31, 2000

Jonathan Katz
U.S. Securities and Exchange Commission
450 5th Street, N.W.
Washington, DC 20549

Re: Securities and Exchange Commission File No. SR-NYSE-99-48, Commission Request for Comment on Issues Relating to Market Fragmentation

Dear Mr. Katz:

I Knight

Knight Trading Group, Inc. ("Knight") welcomes this opportunity to respond to the Commission's Request for Comment on Issues Relating to Market Fragmentation. Knight, headquartered in Jersey City, NJ, is the parent company of Knight Securities, L.P., Knight Capital Markets, Inc., and Knight Financial Products. Knight subsidiaries make markets in equity securities listed on Nasdaq, the OTC Bulletin Board, the New York Stock Exchange and American Stock Exchange, and in options on individual equities, equity indices, fixed income instruments and certain commodities in the U.S. and Europe. The firm also maintains an asset management business for institutional investors and high net worth individuals through Knight Financial Products' Deephaven subsidiary.

As the number one destination for online trade executions, Knight is the processing power behind the explosive growth in securities trading via the Internet. The firm was recently selected to Fortune's "e-50 Stock Index," an elite collection of companies that are shaping the new Internet-based economy. Currently, the five-year-old company employs more than 900 people worldwide with offices in Jersey City, NJ; Jericho, White Plains, Purchase and New York, NY; Chicago, IL; Boston, MA; Minnetonka, MN; Santa Clara, CA; and London, England.

II Summary

The Commission's request for comment is premised upon the issue of market fragmentation. In an attempt to discuss market fragmentation thoroughly, other, equally complex and contentious market structure issues, are invariably addressed by the Commission in its Release. These issues include a laundry list of issues, many of which have been the subject of various past Commission studies, releases or rule-making. These issues include internalization, preferencing, payment-for-order flow, market transparency, conflicts of interest, best execution, and competition between market participants. Knight, in this letter, will provide comment on those issues on which we feel our perspective will add value to the deliberative process.

The Release requests comment on possible alternative regulatory solutions to the perceived ill-effects of market fragmentation that cover a continuum from requiring improved disclosure to mandating order interaction in a central facility. The Commission's proposed solutions to market fragmentation range from those which are largely unobtrusive to existing market structure (requiring increased disclosure) to a re-tooling of the existing structure of the equities markets (mandated central order facility) which would significantly affect, for better or for worse, most existing market participants. At the risk of oversimplification, the Release attempts to ascertain the optimal regulatory framework in which the benefits brought about by competition between market centers and the benefits brought about by competition between orders (price competition) would be maximized.

Knight's comment letter will attempt to persuade the Commission to promote and facilitate competition between market centers which has produced and will continue to produce tangible (i.e. real and not theoretical) benefits to the individual investor-faster, cheaper, more reliable, and higher quality executions. To argue otherwise would belie the facts. "Quote competition" as evidenced by the Nasdaq market before the order handling rules were promulgated or "order competition" as evidenced by the quality of executions on an auction market have simply not delivered what market center competition has delivered. It is the new market centers such as Knight or ECNs that have dramatically improved the quality of executions that the American investor demands. An increasing number of individual investors want immediate executions, not 25-second executions. Individual investors want cheap and reliable executions, not executions that require $300 dollar commissions. Individual investors want securities markets that utilize outstanding advances in technology to deliver increasingly superior products, not to trade in an environment in which orders are entered on slips of paper. Knight, along with a handful of other market participants, has delivered, and continues to deliver, a superior product to the individual investor.

How does one explain that the securities markets today are characterized by innovation, amazing advances in technology, and the delivery of a better and better product to the individual investor? The answer is not because there are merely new market participants such as ECNs and electronic exchanges. The answer is because, for perhaps the first time in this country's history, there is truly fierce competition between the market centers. Therefore, Knight strongly urges the Commission to continue to facilitate the competition that already exists. To dampen or lessen the competitive environment that exists today would significantly harm the individual investor.

III Competition and Technology

Unfortunately, the comment letter process is most often remarkably predictable. On the one hand, market participants who are thriving in the existing regulatory environment will more likely than not offer evidence and support for those alternatives that produce the least amount of regulatory interference or change. On the other hand, those market participants whose business is eroding or threatened by competitive forces will invariably offer evidence to support alternatives that protect their vested interests. While Knight is a market participant whose business is thriving, it would be unfair to cast our comment letter in the "doing well, supports little if any regulatory change" pile of comment letters. That is because Knight is not simply another market participant attempting to protect its interests.

Knight, along with a handful of other Nasdaq market participants, has transformed the way in which individual investors buy and sell securities. Knight is the largest market maker of Nasdaq/Non-Nasdaq OTC Securities and is only five years old. Therefore, Knight is a real-life, not theoretical, example of why the existing market structure has, and will continue, to deliver remarkable benefits to the individual investor. Knight, as well as our competitors, has invested large sums of money in technology to bring a superior product to the individual investor. In fact, we plan to spend over $100 million over the next 12-18 months on technology related investments. We have successfully competed within a regulatory framework that rewards the market center that delivers innovative products to the investor. Improving our product requires the constant commitment of large amounts of capital to enhance our technological capabilities. Many of the Concept Release's alternatives would create significant disincentives for market participants to improve their technological capabilities.

For instance, intermarket time priority for investor limit orders or market maker quotations that improve the NBBO would significantly reduce the incentive for us to improve on the quality of the product that we deliver our customers. If Knight's execution product becomes increasingly reliant upon technologically inferior market centers because of mandated linkages, Knight's competitive advantage will erode and our incentive to continue to commit large amounts of capital to make technological improvements will be reduced. The losers in these scenarios will be those market centers that differentiate themselves through the utilization of expensive and innovative technology, i.e., those market centers that are providing cheaper, faster, and higher quality executions. This, in turn, will significantly and negatively impact the individual investor.

IV Market Fragmentation

The concerns voiced by some market participants that the existence of multiple market centers are creating isolated pools of investor orders, thereby reducing market efficiency, are understandable. However, the concerns about so-called market fragmentation are misplaced. First, it is important to emphasize the following principle: Advancing technology batters down barriers and democratizes access to information. Technology has already produced dramatic improvements in market efficiency-and it will continue to do so. Second, the securities industry's rapid consolidation in recent years has resulted in certain markets that are less fragmented than at any time in recent history. In the Nasdaq market, for instance, the top five market participants now account for about 60 percent of trading volume, compared with 40 percent ten years ago. The mere fact that scores of electronic systems are seeking to become ECNs or ATSs does not signify a fragmented marketplace. As Federal Reserve Chairman Greenspan recently testified: In the long run, unfettered competitive pressures will foster consolidation, as liquidity tends to centralize in the system providing the narrowest bid-offer spread at volume. To a significant extent, this consolidation already has occurred in Nasdaq. It is no coincidence that competition between market centers is clearly most pronounced in those market centers trading Nasdaq securities. The conclusion that we draw from these facts is that fragmentation lessens as unfettered competition between market centers increases. As a result, we strongly urge the Commission to employ its regulatory power to address actual deficiencies, not hypothetical or anticipated deficiencies.

Knight strongly suspects that those voicing the most concern about market fragmentation are really opposed to greater market competition. We also suspect that the call for "market reform" by some of these market participants is really a euphemism for "Let's turn back the clock": Let's turn it back to the days of $300 commissions, slow executions, inefficient technologies, less competition, and less market transparency. Some market participants assume that best execution would be guaranteed in a centralized market or a market in which price/time priority is mandated between market centers. We believe that investors should not have to sacrifice the benefits of an innovative, dynamic marketplace merely to make the fiduciary's job a little easier. Investors are not only focused on the price at which an order is executed. In many cases, speed, enhanced liquidity, and certainty of execution are also important priorities. To eliminate competition, to narrow the markets to a one-size-fits-all approach, would be to pander to the narrow interests of certain market participants, while denying investors many important options.

As a firm that has been built upon supporting market participation by individuals, we should stress that individual investors who send in orders for 300, 400 or 500 shares now provide the Nasdaq market with so much liquidity that even the largest institutional investors are being forced to access that liquidity directly. Some institutions have complained that they are having more difficulty executing their orders--the execution process takes more time, liquidity is sometimes difficult to find, and the way in which larger sized orders are being executed is changing. However these changes are not evidence of fragmentation that support regulatory changes. These changes are evidence that the securities market paradigm is shifting. The market centers processing the order flow from self-directed individual investors are increasingly those providing the highest quality price discovery and the greatest liquidity. The individual investor increasingly controls market prices, not the institutional investor or the institutional brokerage firms.

The issue, then, is not that the marketplace has become more fragmented; it has not. The issue is that liquidity patterns are changing and creating a more level playing field for individual investors. The reason we are having this debate about market structure is that these changing liquidity patterns have eroded the advantage long enjoyed by institutional investors and by the large Wall Street brokerage firms that cater to them. This change is healthy, and it is no reason to stifle competition. In fact, the changes in the marketplace demonstrate that our existing regulatory framework is working and will continue to deliver impressive, tangible results for all investors-individuals and institutions alike.

IV Internalization and Payment for Order Flow

The issues of payment for order flow and internalization have been the subject of incessant debate because they obviously pose potential conflicts: If a market center gives brokers cash rebates or other inducements, how can the investor be sure that an order will receive the best possible execution? Knight is in the position of defending a practice that costs it an enormous amount of money. Last year, we rebated nearly $139 million to our broker-dealer clients who sent us their order flow. What our client firms did with that money is a question best answered by them, but no doubt they would tell you that, because of those rebates, their customers paid lower commissions, received more free real-time market data, and more free technical and fundamental analysis on more and more securities.

The securities industry is rife with conflicts. Besides payment for order flow, there exists a conflict when the broker and dealer functions is combined in one company; or when securities firms both underwrite securities and issue research reports on them. Perhaps the most conspicuous conflict on Wall Street is that broker-dealers habitually recommend the purchase of securities for which they also make markets. The solution to such conflicts is not to disallow them-which would do damage to the efficiency of our capital markets. Rather, it is to insist upon full and fair disclosure.

Payment for order flow should be approached in a similar light. Indeed, the SEC and Congress debated the issue in the early nineties and concluded that full and fair disclosure was the right approach. By giving investors high-quality, real-time market information and execution results, by providing them with a clear explanation of each market center's policies, procedures, execution standards and trading protocols, and by fully disclosing any practice that might be considered a conflict of interest, regulators and market participants will enable the individual investor to make intelligent choices among all of the available options. Outlawing the practice of payment for order flow would lessen the ability of investors to access immediate liquidity for any securities they wish to trade. Rather than denying investors this access simply because of a potential conflict, it would be far better to ensure that they have the information and the tools they need to detect evidence of inferior service.

There is strong evidence that cash rebates are not adversely affecting firms' order routing decisions. For instance, some firms that do not accept cash rebates route their customer orders to Knight and receive the same quality executions as firms that do accept cash rebates. The real power that brokers possess is not the ability to demand cash rebates, but rather, to demand higher quality executions for their customers. On-line brokers are able to assert the power of many with the voice of one. On-line brokers constantly demand that we find innovative ways to improve the quality of our executions for their customers so that they, in turn, are able to attract new investors and retain their existing accounts. This competitive pressure has resulted in guarantees that would have been unheard of a few years ago, such as, guaranteed automated price improvement, guaranteed executions greater than the quoted size in a matter of seconds, mid-point pricing of incoming market orders with opposite side market orders, and mid-point pricing at the opening of the market and IPOs.

Cash rebates are not the only form of inducement for order flow. To consider this issue fully, one also needs to examine internalization, reciprocal order-routing arrangements, clearing arrangements, soft dollars, directed brokerage, payments by ECNs to add liquidity, and commission recapture programs-where institutions and their advisers receive rebates in the form of cash, research, or brokerage services in direct relation to the amount of order flow sent to a broker-dealer. The accepted broker-dealer practice of executing orders internally poses the same potential conflict. In each case, the solution, in our opinion, to potential conflicts lies in full and fair disclosure.

VI Market Transparency

The regulation of U.S. securities markets-arguably the most enlightened in the world-has always rested on the precept of full and fair disclosure. That is, by providing all investors-not just market professionals-with high quality information, we empower them to make intelligent choices among all the available options without unnecessarily stifling competition or innovation.

Chairman Levitt's call for a voluntary private sector initiative to make limit order books available to the public and Commissioner Laura Unger's recommendation that all market centers be required to provide uniform information on various best execution factors should be adopted. Both of these recommendations would improve market transparency. Better-informed investors are more likely to participate in the capital formation process, making our securities markets even deeper and more liquid in the process.

There is no justification for the age-old practice of market professionals walling off access to important market information. It is neither fair to individual investors, nor healthy for the marketplace. All investors should be provided not only with quotation and trade information, but also with the limit order files of the various market centers and with information that objectively measures the quality of execution at those market centers.

Chairman Levitt's recent call for a voluntary private sector initiative to improve the transparency of limit order books across all markets is vital to ensuring that investors receive the benefits of intermarket competition. Knight wants to ensure that limit order information becomes available to the individual investor, and we look forward to working with the SEC and other securities market participants to that end. Commissioner Unger's suggestion is equally important. If investors themselves had the information to identify which market centers have superior execution standards and trading protocols, the benefits of competition between market centers would be better realized. At Knight, we have begun publishing our execution standards and protocol on our website. We will soon begin publishing execution results as well--including price improvement rates, enhanced liquidity rates, immediacy rates, and price disimprovement rates--and we encourage all of our competitors to do the same.

VIII Ill-Effects of Market Fragmentation is Reduced through Self-Policing

Knight believes that many of the concerns outlined in the Concept Release could more efficiently be addressed by further empowering the fiduciary or the individual investor. Mandating price/time priority is a significant regulatory endeavor that, as discussed above, may have significant deleterious effects. However, if fiduciaries or the individual investor rigorously monitored the various market center alternatives, and actively routed orders to those market centers that provided the best execution, the need for burdensome regulations would be mitigated.

For instance, the Commission, in its Concept Release, expresses concern that a customer limit order, under the existing regulatory structure, could remain unexecuted in one market center as dozens of orders at the same price, entered later in time, are executed on another market center. Knight believes that this scenario does not highlight a regulatory weakness as much as it highlights a lapse in the fiduciaries' obligation to route its customer orders to those market centers that provide the best possible execution.

In this scenario just described, best execution would be measured by the likelihood of execution of a customer limit order. If a fiduciary routes its customers' limit orders to an illiquid ECN or to a market center that receives a small amount of order flow in a particular security, the likelihood of execution is, not surprisingly, much lower than if the order were routed to a market center receiving a large volume of market orders (as the market orders are matched with the limit orders) or to a liquid ECN. If fiduciaries vigilantly monitored the execution quality of the various market centers, as they are required to do, and actively routed their customers' orders to those market centers that provide the best possible execution, the ill-effects of market fragmentation would be significantly reduced.

IX Conclusion

In conclusion, Knight believes that by promoting vigorous competition among competing market centers, our existing market structure has not only adapted to changing market patterns; it has delivered spectacular results for the individual investor. It has promoted unprecedented innovation and service for investors. It has helped to revitalize our markets and preserve their status as the envy of the world.

Lastly, please incorporate, by reference, the Congressional testimony given on April 26, 2000 before the Subcommittee on Securities of the U.S. Senate Committee on Banking, Housing and Urban Affairs of Knight's President and CEO, Kenneth Pasternak. If I can be of further assistance to you on this matter, please do not hesitate to contact me at (800) 544-7508.



Michael T. Dorsey
Senior Vice President and
General Counsel