Hearing on "Maintaining Leadership in the Financial Marketplace of the Future"

Written Statement of

David G. Downey
Executive Vice President, Interactive Brokers, L.L.C.

9:00 a.m., Monday, May 8, 2000
3rd Floor Conference Center - Federal Reserve Bank of Chicago
230 South LaSalle Street, Chicago, Illinois


Chairman Gramm and members of the committee my name is David Downey and I am Executive Vice President and Head of Operations at Interactive Brokers LLC. Interactive Brokers is a registered Broker/Dealer and Futures Commission Merchant and is a member of the Timber Hill Group of Companies, which along with its affiliated entities, are members of most commodities and securities exchanges in the United States, and several foreign exchanges - in all more than 30 exchanges in 15 countries on 4 continents. More than five percent of the exchange based equity options and futures traded each day worldwide are traded through the Interactive Brokers network.

On behalf of Interactive Brokers and The Timber Hill Group, I respectfully submit this testimony regarding the critical market structure issues facing Congress, the Securities & Exchange Commission, and the securities industry. My testimony raises the same themes as a comment letter Interactive Brokers' recently filed with the SEC in response to the Commission's recent concept release on market fragmentation issues, and I am grateful for the opportunity to address these important matters before this committee.

I. Introduction and Summary

At the outset I would like to thank this committee as well as the SEC for taking the initiative to call for public debate on the critical market structure issues that have arisen with the proliferation of trading venues and the rapid advancement of technology for securities trading. On these issues, policy makers are asking the right questions at the right time, and we are hopeful that Congress, the Commission, and the public can work toward a market structure that will capture the benefits of multiple, competing marketplaces while at the same time providing customers with best execution of their orders notwithstanding a decentralized trading environment.

In this testimony, we propose certain market structure elements that would constitute a workable resolution of the various controversies surrounding "market fragmentation." Unfortunately, public discussion of this issue recently has become a somewhat sterile debate between those who favor a national central limit order book ("CLOB") based on price/time priority, and those who argue that Congress and the Commission should have no role in addressing the increasingly troublesome issues arising from "step-up" and match guarantees, internalization, payment for order flow and other practices that have the potential to subvert price competition and deny public customers true best execution of their orders. As described fully below, our proposal is a compromise that would duplicate the competitive benefits of a central limit order book (i.e., rewarding market participants for displaying more competitive prices, increasing transparency of trading interest, preventing trade-throughs, and ensuring that customers get best price execution), while at the same time avoiding the serious drawbacks that would come from a government-mandated, centralized "super-exchange" (i.e., diminished incentives for separate market centers to compete to provide innovative products and services, susceptibility of a single, monolithic system to delays, outages and capacity constraints, difficulty of administration and resistance to change or improvement once entrenched, etc.). Moreover, our proposal relies on the enhancement and evolution of existing market mechanisms, rather than the creation of a host of new rules and regulations and a new linkage bureaucracy that Congress and the Commission would be tasked to oversee.

Simply stated, we propose that it be made clear that broker-dealers, pursuant to their existing duty to provide best execution of customer orders, bear the responsibility to route each customer order to the best posted market (or markets) for a particular security. Broker-dealers required to ensure order-by-order best execution will naturally come to satisfy this duty by surveying the limit order books for each available market (which would be made available in their entirety) and compiling a "locally-aggregated limit order book" ("LALOB") that would contain information on all available trading interest at every major market where a particular security is traded. Each broker dealer would use the information in its locally-aggregated limit order book to route each customer order to the market posting the best price (or to multiple markets if no single market was posting sufficient size at the National Best Bid or Offer ("NBBO") to execute the customer's order). To encourage price competition and to ensure that each market's prices would be fully transparent to broker-dealers, the Commission should reaffirm that its "Quote Rule" (11Ac1-1) requires all market makers and specialists to post the actual best prices at which they are willing to trade, and that all quotes or guarantees to trade therefore must be posted as firm, executable orders accessible to all market participants. Market centers thus would not have the right to guarantee to "step up" and match a better price posted at another market, while posting an inferior price at their own market.

Relying on broker-dealers to route individual orders to the best posted market based on the broker-dealers' locally-aggregated, inter-market limit order books would provide all the benefits of a CLOB without any of its shortcomings. Customers would get best execution of their orders free of the corrosive influence of internalization or payment for order flow. Public investors and professional liquidity providers would have a strong incentive to post better prices and narrow the spread, since broker-dealers would have to route orders to the best posted markets and market makers would not be able to guarantee to step up and trade at prices that they were unwilling to post publicly. Moreover, this all would be accomplished by reaffirming and enforcing the Quote Rule and broker-dealers' existing fiduciary duties. No monopoly linkage authority would be created that could subsume existing exchanges and ECNs or eliminate their incentive to develop innovative products, services and technologies. Perhaps most importantly, a market "linked" by a broad network of broker-dealers, each compiling a LALOB and routing orders accordingly would -- like the Internet itself -- be highly redundant and have tremendous capacity. There would be no single point of failure or capacity constraint and system or computer security problems from a particular market center or broker-dealer would not impair the system as a whole.

Multiple market centers, competing vigorously for order flow based on firm, posted prices, offer the best hope for reducing explicit and implicit trading costs borne by investors. The proposal outlined herein would mitigate the potential problems arising from these "fragmented markets" but would do so by harnessing competitive, free-market forces, with minimal additional need for governmental interference or oversight. As such, we think it offers the best compromise of the positions being urged by the various parties to this debate.

This testimony first discusses some of the problems arising from the current structure of the securities markets (namely, exchange "step-up and match" guarantees that undermine market transparency and price competition; and ineffective linkages between markets). We then address the drawbacks of the solutions commonly proposed to these problems. Finally, we set forth in detail our proposal for market structure enhancements based on broker-dealer best execution routing and the use of locally-aggregated limit order books.

II. Concerns Regarding Price Competition, Transparency
and Best Execution Are Well-Founded.

A. Step-Up and Match Guarantees Undermine Price Competition and Market
Transparency and Violate the Quote Rule

In order to eliminate price competition from other trading venues, exchanges increasingly are implementing policies under which market makers or specialists may guarantee to step up and match more competitive prices posted by other markets in order to fill incoming customer orders. Thus, even if an exchange initially receiving a customer's order was not posting the best price in the national market, market makers on that exchange would have the right to execute the trade at the NBBO. As the SEC recently explained:

"[T]he market center to which an order is initially routed is permitted to match the best price and execute the order internally. Indeed, the executing market center need not ever have displayed the best price. [T]he market participant (whether investor or dealer) who publicly displays an order or quotation at a better price than anyone else is offering is not entitled to any assurance that the order or quotation will interact with the next trading interest on the other side of the market."

Id. at 10583. Step-up and match policies widen markets by removing the incentive for different market centers to compete on price, they erode the transparency of the National Market System, and they are contrary to Commission Rule 11Ac1-1 (the "Quote Rule").

The right to step up and trade at others' prices creates a strong incentive for market makers to post worse prices, creating wider markets. For example, a market maker lowering his bid increases the chance that other, competing exchanges, will also lower their bids and that whichever exchange gets the next customer order to sell at the market will be able to pay a lower price to that customer, garnering a larger profit. On the other hand, if not all exchanges follow the market maker's lower bid, nothing is lost because the exchange receiving an order simply may step up and match the better away price. Thus, a market center that quotes a wide market increases its chance of profiting from trading that wide market, without any negative consequences. 1

In economic terms, step-up and match guarantees create a problem of negative selection. It is not in the best interests of a dealer to quote a firm, narrow market if other dealers with order flow can simply step up and match that dealer's posted prices when they like an incoming order, but pass incoming orders that they do not want to the quoting dealer (who, unlike the dealers not posting the tight market, must abide by his firm quote obligations and execute the trades). This creates negative selection in which other dealers cherry-pick the order flow and the dealer quoting the best market ends up only with the potentially unprofitable trades that others did not want.

In a market structure that allows market makers freely to promise to step up and match better prices posted by others, market makers will not compete based on price but based on their ability to internalize orders or pay for order flow. As the SEC notes in its market fragmentation concept release: "[A] market maker with access to directed order flow often may merely match the displayed prices of other centers and leave the displayed trading interest unsatisfied. The profits that can be earned by a market maker trading at favorable prices with directed order flow can then be shared with the brokers that routed the orders." 65 Fed. Reg. at 10583. The Commission goes on to ask whether "it [is] possible for a non-dominant market center to compete successfully for order flow by price competition, without using internalization and payment for order flow arrangements?" 65 Fed. Reg. at 10586. 2 In a system where specialists and market makers can step up and trade with captive order flow at the NBBO, even if they do not care to post that NBBO, the answer to the Commission's question is "No."

In addition to impairing price competition and rewarding market makers for posting wide markets, step-up and match guarantees fundamentally undermine the transparency of the national market system, and are contrary to the SEC's Quote Rule. Congress and the Commission have stated repeatedly that price transparency is the bedrock upon which fair and open securities markets rest. See e.g., id. at 10583 ("price transparency is a minimum essential component of a unified national market system"). To this end, the Quote Rule requires specialists and market makers "to promptly communicate to [their] exchange . . . [their] best bids, best offers and quotation sizes," and requires exchanges to post those best bids and offers with associated size. 3 Yet with step-up and price matching, market makers routinely trade at prices that they do not post, and the prices that they do post do not reflect the actual best prices at which they are willing to trade.

The situation created by step-up and match guarantees is very much like the situation that the Commission sought to address when it amended the Quote Rule in 1996 to require specialists and OTC market makers to post publicly the price of any order they place at an ECN if that price is superior to the price posted by the specialist or market maker on its own market. The Commission was concerned that specialists and market makers were posting artificially wide markets and then offering better prices, away from public view, on ECNs. As the Commission explained:

"[T]he Commission for many years has been concerned that the development of so-called "hidden markets," in which a market maker or specialist publishes quotations at prices superior to the quotation information it disseminates on a general basis, impedes [National Market System ("NMS")] objectives. Over the course of the last decade, certain trading systems that allow market makers and specialists to widely disseminate significant trading interest to certain market participants without making this trading interest available to the public market at large have become significant markets in their own right. Although offering benefits to some market participants, widespread participation in these hidden markets has reduced the completeness and value of publicly available quotations contrary to the purposes of the NMS."

Exch. Act. Rel. 37619A, 62 SEC Docket 2083 at 2112-2113.

Likewise, allowing specialists and market makers routinely to trade at prices that they are unwilling to display publicly creates a hidden market in which public investors have no way to determine what the real best price for a security is or how much liquidity is available at that price. As set forth in our proposal below, the Commission should clarify that the Quote Rule means what it says - that market makers and specialists must post the actual best prices at which they are willing to trade and that all quotes or guarantees to trade must be posted as firm, executable orders accessible to all market participants. Market centers posting inferior prices would not have the right to guarantee to "step up" and match better prices posted on other markets. Thus, broker-dealers routing orders to those market centers with inferior prices would not be acting in accordance with their fiduciary duty.

B. Current Market Linkages Are Ineffective and Customers Often
Do Not Get True Best Execution of their Orders.

In addition to concerns about market transparency and price competition, Congress and the SEC are right to be concerned about the effectiveness of existing market linkage mechanisms, and about whether investors are getting true best execution of their orders. Many commentators have outlined the inadequacies of the ITS. It is technologically obsolete. It does not offer automatic order execution. And execution of orders sent through the linkage is slow and uncertain at best. Further, as is the danger with any monopoly operating under public auspices, the participants in ITS -- by manipulating its governance structure -- have used it to maintain their market positions and exclude competition by denying new market centers access to the linkage. Likewise, the options markets have been unable or unwilling to devise an effective market linkage plan.

Broker-dealer systems, and not a centralized exchange linkage, should be the mechanism through which customer orders are routed to the best market. A broker-dealer's duty of best execution "requires the broker-dealer to seek to obtain for its customer orders the most favorable terms reasonably available under the circumstances." See e.g., Newton v. Merrill Lynch, 135 F.3d 266, 270 (1998). Congress, the courts and the Commission have recognized that the standard imposed on a broker-dealer in finding the best terms for its customer - normally the best price - becomes higher as technology advances. See 61 Fed. Reg. at 48306-16; Newton at 271. Yet to date, although they have a fiduciary duty to execute their customers' orders on the most favorable terms available, and although best execution order routing systems can be easily developed or purchased, many broker-dealers do not route each customer order individually to the best market, instead defaulting to long-standing relationships or accepting payment for order flow and arguing that best execution must be viewed "holistically" or "in the aggregate" rather than customer-by-customer and order-by-order. As set forth fully below, policy makers should make clear what has been implicit for some time: that our markets and the technology to access those markets have evolved such that the duty of best execution requires individual, order-by-order routing of customer orders.

III. Shortcomings of Certain of the Proposed Solutions to
Market Fragmentation Concerns

A number of proposals have been put forward to address issues stemming from market fragmentation. While elements of each of these proposals have merit, they either would not address the full scope of the issues, or their adoption would create significant new problems or require unwieldy new rules and regulations.

A. Disclosure of Internalization and Payment for Order Flow.

Some commentators have argued that the regulators should do no more than require some form of enhanced disclosure regarding broker-dealer order routing practices such as internalization and payment for order flow. See 65 Fed. Reg. at 10586. We respectfully suggest that disclosure is not sufficient in and of itself to address the serious market distortions arising from these practices.

Unless disclosure of payment for order flow and internalization arrangements caused a wide segment of the investing public to cease to deal with broker-dealers who were party to these arrangements - a very dubious assumption - disclosure would do nothing to alter the powerful disincentive to price competition resulting therefrom. Indeed, there is little reason to believe that disclosure of order routing practices would do much to change the way order flow currently is determined. Internalization and payment for order flow arrangements can be very complex, and will not be susceptible to easy, comprehensible explanation. Payment for order flow often involves hidden or indirect incentives that will not be readily apparent to the reader of a boilerplate disclosure. Likewise, broker-dealers may have numerous order-routing arrangements, covering different products under different circumstances.

The very purpose of internalization and payment for order flow schemes will be hard for many members of the investing public completely to fathom. As the Commission notes in its recent concept release, even many sophisticated investors are unaware of the "implicit costs" they pay by giving up the bid-ask spread when using market orders or marketable limit orders. Id. at 10580-81. Seemingly few investors understand that their order may have profit potential to their broker (and the market receiving the order) far in excess of the commission they are paying. Ultimately, order routing disclosures likely would either be so generalized and non-specific as to be essentially meaningless, or else they would be so detailed and byzantine as to make their review and comprehension by even sophisticated customers highly unlikely. Either way, there is little reason to expect that order routing disclosures would have a substantial impact on customers' choice of brokers, or would even slightly mitigate the anti-competitive effects of the practices being disclosed.

B. Restricting Internalization and Payment for Order Flow.

Some have suggest that the SEC restrict internalization or payment for order flow, for example by adopting the proposal of the New York Stock Exchange ("NYSE") under which a broker-dealer would not be able to step up and match the NBBO to trade with a customer order, but would have to offer price improvement over the NBBO to do so. Id. at 10586. Again, this would not solve the problem. Market transparency requires that broker-dealers only trade at prices that they have publicly posted, as they are required to do under the Quote Rule. If specialists and market makers were required to post their true best prices, and broker-dealers routed orders to the best posted markets, internalization and payment for order flow - whether or not they persisted -- would cease to be of concern because they would no longer impair price competition among markets or affect best execution of customer orders.

Under the NYSE proposal, it would still be impossible to determine the true best price for a security or the liquidity available at that price because dealers could still step-up and trade at undisclosed prices so long as they offered price improvement. Dealers could still maintain artificially wide markets, and then offer modest "price improvement" by stepping up to better the NBBO (e.g., by offering 5 cents' price improvement in a 75 cent-wide market).

C. Requiring Exposure of Market Orders to Price Competition

Another proposed solution to market fragmentation concerns is to require market centers to expose their market orders and marketable limit orders publicly in some manner for the purpose of obtaining price improvement. Id. at 10586-87. Again, we think this would offer only a partial solution to the relevant problems and would be unnecessary if the proposal discussed herein were adopted.

As noted above, "price improvement" in a wide market can be something of a red herring and should not substitute for a market structure that fosters the narrowest possible spreads. If market makers and specialists could only trade against market orders and marketable limit orders if they were quoting at the NBBO before an order was received, market makers would have to post or better the NBBO in order to attract order flow. This would create the narrowest markets, with the greatest posted size, for all investors, and there would be a very high likelihood that a market center executing against an incoming market or marketable limit order would be doing so at the best possible price. In addition to being technologically cumbersome, devising mechanisms to try to achieve price improvement would be unnecessary because publicly posted market prices, available to all comers, already would be "improved".

D. Adopting an Intermarket Prohibition Against Market Makers Trading Ahead of
Previously Displayed Customer Orders.

In its concept release, the Commission also called for comment on whether to "adopt an intermarket prohibition against market makers (including exchange specialists) using their access to directed order flow to trade ahead of investor limit orders that were previously displayed by any market center and accessible through automatic execution by other market centers." Id. There are a number of flaws with this approach.

Implementation of intermarket customer time priority would be difficult and would require a system to be constructed so that each market center would know whether the prices posted by another market center represented a customer limit order, and who was first in time priority - along with requiring an intermarket automatic execution facility. Building and operating such a system would be close to or as difficult as building and maintaining a national central limit order book.

In addition, an intermarket prohibition against market makers trading ahead of prior investor limit orders would require the adoption of a complaint and satisfaction mechanism that would likely be ineffective. In fast moving, increasingly automated markets, many customers or their brokers either will not notice or will not timely complain about violations of the prohibition, because of the time and effort involved to do so. Market centers therefore will know that they can profit from whichever violations go unmentioned or untimely reported, while simply paying back any extra profits earned from violations that actually result in a timely complaint.

Finally, while offering a priority to public customers may make step-up and match guarantees seem somewhat less unpalatable to policy makers, it will do nothing to address the greater problem - the lack of incentive for specialists and market makers to post better prices. Although customers often may narrow the spread and should be rewarded for doing so (as they would be under our proposal), this is not yet a substitute for the liquidity provided by professional market makers seeking executions. See id. at 10582 ("one of the most significant benefits of providing an opportunity for multiple dealers to participate in the national market system (often through competing market centers) is provided by their willingness to step in and supply liquidity at prices that will absorb temporary imbalances in the trading interest of investors."). 4

E. Implementation of a Central Limit Order Book Based on Intermarket
Price/Time Priority Would Create a Host of New Problems.

Creation of a national linkage system (or central limit order book) based on price/time priority is the most far-ranging and controversial of the proposals that have been outlined to address market fragmentation. While price/time priority principles are preferable to a step-up and match/payment for order flow model, we think that Congress, the Commission and the industry would be misplaced in placing the primary responsibility for order routing on a single, inter-exchange linkage system, rather than on the broker-dealers who currently receive and route customer orders.

A national central limit order book or "super-exchange" would reduce existing market centers to mere entry portals into the system. It is difficult to imagine the continued viability of multiple, competing marketplaces in such a system, or how those marketplaces would have the incentive or resources to develop new products, technologies or services. Indeed, notwithstanding the market fragmentation issues presented because multiple exchanges trade the same products, there are many benefits to the continued operation of multiple, independent markets. Aside from price competition, there is greater competition to develop new products (e.g., like the standardized stock options pioneered by the CBOE or the depositary receipt products pioneered by the Amex), and to list new and existing products.

The existence of multiple, independent market centers also provides greater redundancy and capacity to the National Market System, so that if the resources of one market are taxed or its services disrupted, trading can continue on other exchanges. Even with respect to their self-regulatory functions, having multiple exchanges with different approaches to regulatory oversight can foster innovation and the development of better and wiser exchange rules and policies.

On the other hand, a monolithic inter-exchange linkage through which all or substantially all customer orders would be routed in price/time priority would present all the problems of any regulated monopoly. The system would have a single failure point and would be susceptible to delays and outages. The system would be inflexible and resistant to change. With existing exchanges presumably merging or disappearing altogether there would be little incentive for the operator of the linkage to be responsive to the demands of members or customers because there would be nowhere else to trade. And our experience with ITS suggests that once entrenched, it would be very difficult for new and innovative trading venues to be established to compete with the exchange or exchanges that control the linkage system.

Another problem with a central limit order book linking existing markets is that it may be a less comprehensive solution than it seems. For example, foreign markets (trading either U.S. or non-U.S. securities) presumably would not be encompassed by such a regime. In addition, new trading venues and mechanisms beyond traditional exchanges and ECNs may arise over which there may be gaps in the SEC's rules or rulemaking authority. A "linkage" system grounded on the fiduciary obligations of Commission-regulated broker-dealers offers a more comprehensive solution to market fragmentation concerns.

In sum, while we fully support the principle of intermarket price/time priority, we believe that the goals of a national central limit order book can be achieved more comprehensively, and with fewer negative side effects, by relying on broker-dealers to route orders to the best posted markets, as described below.

IV. Best Execution Order Routing Is the Most Workable and Least Intrusive
Solution to Concerns Arising from Market Fragmentation.

Broker-dealers, pursuant to their existing duty to provide best execution of customer orders, should bear the responsibility to use systems (or arrange to access systems) that route each customer order to the best posted market (or markets) for a particular security. By simply: 1) clarifying that a broker-dealer's existing fiduciary duty to its customers includes the duty to route each order individually; and 2) reaffirming that the Quote Rule means that market makers and specialists must post their true best prices; Congress and the Commission can create the same powerful incentives for increased liquidity, transparency and price competition as an intermarket price/time priority system or CLOB, without any of the drawbacks of a centralized linkage.

Congress, the Commission and the federal courts have stated repeatedly that broker-dealers' are expected to take advantage of all "readily available" technology to ensure best execution of customer orders, and that the standard imposed on a broker-dealer in finding the best terms for execution of its customer orders becomes higher with advances in telecommunications and technology. See e.g., 61 Fed. Reg. at 48306-16; Newton v. Merrill Lynch, 135 F.3d 266, 271. As Chairman Levitt has pointed out, "systems for broker-dealers recently have emerged that include sophisticated algorithms for automatically routing investor orders in a security to the best market." 5These systems can be built, acquired or utilized by broker dealers at increasingly low cost, and in an age where even relatively unsophisticated computer users can use their PC's to find the best price for books, CD's or airline tickets, there is simply no excuse for broker-dealers not to use computerized systems to route customer orders to the best markets.

Given the current state of technology, broker-dealers required to satisfy their duty of best execution on an order-by-order basis would naturally do so by accessing each available market and compiling a locally-aggregated limit order book ("LALOB") that would contain all the information from the limit order books of each major market center where a particular security is traded. Each broker dealer would use the information in its LALOB to route each customer market order or marketable limit order to a market posting the best price. If no single market was posting sufficient size at the NBBO to execute the customer's order, the broker-dealer would divide the order across markets in order to optimize aggregate execution price and speed. If another order that exhausted the volume available at a posted price traded ahead of an order subsequently sent by a broker-dealer seeking that price, the broker-dealer that sent the subsequent order would cancel it and reroute it accordingly.

To achieve these results, it should be reaffirmed that the Quote Rule means exactly what is says - that broker-dealers must publicly post the actual best prices at which they are willing to trade. If market centers posting inferior prices cannot guarantee to "step-up" and match better prices posted at other markets, lest they violate the Quote Rule, broker-dealers will be able to attain best execution only by routing each order to the market posting the NBBO. Unlike the current system, this will provide a compelling incentive for a market center to narrow the spread and be at the best price before a broker-dealer makes its routing decision (so that the market center gets the order). See 65 Fed. Reg. at 10580 ("Price competition also may be enhanced by competition among market centers when this involves multiple dealers competing for order flow based on displayed quotations."). Indeed, even if this reaffirmation of the Quote Rule merely caused specialists and market makers reflexively to post any better price posted by their competitors, that would have a profoundly beneficial effect on the markets. The full extent of liquidity at the NBBO, across the National Market System, would be revealed and accessible to brokers, who would thus be able to buy and sell even very large amounts of securities for their customers at the NBBO.

Under our proposal, a broker-dealer would not have to route a customer order to the market center that was first at the NBBO, but could choose among market centers posting the NBBO when the broker-dealer routes the order. Price priority alone will create incentives for a market to better the NBBO and for other markets to add posted liquidity to the NBBO in order to compete for orders. Requiring broker-dealers to track the time priority of various markets would be burdensome, would create an extra layer of message traffic, and would lead to disputes regarding who truly was first at a given price (necessitating creation of a potentially-unwieldy complaint and satisfaction mechanism - see above). Perhaps most importantly, allowing broker-dealers to route orders to any markets publicly posting the NBBO, regardless of time priority, will allow those markets to compete based on size and other attributes such as technology and quality of service.

Broker-dealers would be able to route non-marketable limit orders to any market trading a given product since, regardless of its original destination, a non-marketable limit order would appear on the LALOBs of other broker-dealers and would be equally transparent to the National Market System while resting on any exchange's book. Market centers would provide automatic execution of market and marketable limit orders, at least to the extent of their minimum quote size, proceeded by customer limit orders at the top of the book.

Interactive Brokers has developed a best execution order routing system but we are by no means alone in adopting this technology, nor are we alone in urging the use of best execution systems to address market fragmentation concerns. In just the last several weeks, a report by a committee of the outside directors of the New York Stock Exchange regarding the ITS ("NYSE Committee Report") was made public, calling for the elimination of the ITS and calling for broker-dealers to route each individual order to the best market:

"[Developments in communications technology have eliminated the need for an intermarket order-routing system such as ITS Given these advances, we believe that broker-dealers now have the ability to fulfill their fiduciary obligations to deliver best executions on an order-by-order basis without the need for intermarket order-routing linkages."

See "NYSE Panel Calls for Elimination of System Linking U.S. Markets," Wall Street Journal (Apr. 7, 2000).

The solution outlined herein rests firmly on existing market principles. All it would require is that the Commission make explicit broker-dealers' order-by-order routing obligation and reaffirm that market makers must post their best quotes or guarantees to trade as firm, executable orders. It would not require the wholesale creation of a new linkage system with an array of new rules and oversight and governance procedures. Nor would it undermine or threaten existing exchanges or ECNs or deprive them of any incentive for continued competition. A national market system linked by a broad network of broker-dealer order routing systems would also be highly robust and resistant to outages, delays and computer security threats.

V. Conclusion

Relying on broker-dealers to route orders according to their survey of the prices posted on available markets is the simplest, surest way to ensure that customers get best execution of their orders and that multiple market centers have an incentive to compete with each other to provide the best prices for execution of those orders. The proposal herein has the added virtue of offering a workable compromise of the strongly-held positions on both sides of the debate.

Again, we commend the SEC and Congress for engaging the industry and the public on these issues.


1 -- This is why guarantees among exchanges to trade at a single national price would constitute a violation of the antitrust laws absent Commission approval immunizing them from liability. See e.g., Falls City Industries, Inc. v. Vanco Beverage, Inc., 460 U.S. 428, 441 (1983) (citing FTC v. A.E. Staley MFG Co., 324 U.S. 746 (1945) (setting prices according to a single scheme by its nature precludes independent pricing in response to normal competitive forces and is therefore illegal)).

2 -- Indeed, in response to the competitive pressures arising from multiple listing of options, the Chicago Board Options Exchange has devised a major new payment for order flow initiative. See "CBOE Trader Group Unveils Plan to Install System for Buying Orders from Retail Brokers," Wall Street Journal (Apr. 6, 2000)(" The CBOE association intends to concentrate on attracting retail options orders that can almost always be filled at the bid or offer, which enables dealers to pocket the spread - the difference between the two prices.").

3 -- 17 C.F.R. 240.11Ac1-1 (emphasis added).

4 -- Another problem with an intermarket priority for "customers" is that it is very difficult to define that term and to delineate those who may deserve trading preferences from those that do not. It is not uncommon for broker-dealers to give up their licenses so that they can avoid restrictions on broker-dealer trading and take advantage of preferences for customers. Likewise, "customers" may be hedge funds, proprietary trading firms or one of many other varieties of trader for whom there may be scant public policy justification to grant an advantage over exchange members.

5 -- Hearing Before the Senate Subcomm. on Securities, Comm. on Banking, Housing, And Urban Affairs Concerning Market Structure Issues Currently Facing the Commission (Oct. 27, 1999)(statement of Chairman Arthur Levitt).

Home | Menu | Links | Info | Chairman's Page