Letter From William Brodsky To Arthur Levitt Regarding Options Exchanges Linkage

Dear Arthur: CBOE is writing in response to your November 10, 1999, letter requesting our views on whether to incorporate price-time priority into an intermarket options linkage. We appreciate your invitation to comment on the issue, which has engendered much discussion among the options exchanges and within the broader options industry. An effective options market linkage is an important objective and we want to commend you for your leadership in this area. CBOE likewise has been in the forefront of recent options market initiatives during the past year, including a reduction in customer transaction fees, the introduction of an automated opening rotation system and the expansion of multiple trading this past August. In addition, pursuant to your request, we developed a comprehensive linkage proposal for consideration by the options exchanges to address concerns about the quality of investor executions in the expanded multiple trading environment. CBOE is intensely committed to pursuing initiatives to keep the U.S. options markets the most competitive and the fairest in the world for investors. That is why we feel so strongly that the application of price-time priority as the sole component in an intermarket option linkage is wrong - it is bad for the options market and investors and is a diversion from formulating a real linkage plan. CBOE takes seriously the SEC Order of October 19, 1999, directing the options exchanges to develop a joint linkage plan for multiply-traded options ("SEC Order"). Like you, we agree that with the recent increases in multiple trading, a linkage is needed to prevent trade-throughs and ensure best execution of customer orders. CBOE is committed to implementing a real linkage and has put forth a proposal that addresses all of SEC's concerns. The SEC Order listed the specific items the SEC believes the linkage plan needs: (1) uniform trade-through rules; (2) expansion of the public customer definition to include agency orders presented by competing exchanges; (3) repeal of &trade or fade& rules at the options exchanges; (4) a firm quote requirement; and (5) transmission of orders between options exchanges on a nondiscriminatory basis. CBOE's linkage plan would enhance best execution opportunities for customers by requiring that each exchange adopt an intermarket trade-through rule which would restrict trades from taking place outside the National Best Bid or Offer ("NBBO"). To facilitate compliance with the trade-through rule, an intermarket order routing system would be implemented so that orders on behalf of customers and market maker orders up to a certain size could be routed between exchanges and receive automatic executions. Thus, if an exchange initially receiving an order were not willing to provide the NBBO price to an order, it could access the exchange disseminating the NBBO price and receive an automatic execution at that NBBO price. In this way, a firm would continue to be able to route its customer orders to the exchange of its choice, assured that the receiving exchange could execute the order at the NBBO or better. The CBOE plan also would entail expanded firm quote requirements and requirements that exchanges display the markets of all competing exchanges at the location where multiply traded options are located. "Trade-or-fade" rules would largely be eliminated under our plan. Finally, CBOE's linkage proposal is feasible and can be implemented quickly and efficiently using recently developed technology that is currently being used in the industry. In sum, our plan addresses precisely what the SEC requested and what investors need. In your November 10 letter, you ask CBOE for its views on a concept - price-time priority - that is far outside of a market linkage system. Although the price-time priority feature is not specifically described in your letter, we assume it means a mandatory routing requirement where orders would be required to be routed automatically from a broker-dealer to the exchange that posts a quotation first in time at the NBBO. As we understand it, if an exchange were to receive an order and was not quoting at the NBBO when it received the order, but was willing to match the NBBO, it would be prevented from matching the NBBO and would be required to redirect the order to the exchange that was first quoting the NBBO. In addition, even if an exchange received an order while it was quoting at the NBBO price, if it were not the first exchange to have quoted at the NBBO, it would be required to redirect the order to the exchange that quoted at the NBBO first. We refer to this feature as a mandatory routing switch because it automatically forces a customer's order to be routed to a particular exchange. As described more fully in this letter, we believe that a mandatory routing switch would have a destructive impact on investors and the options market, is contrary to the Commission's prior initiatives in fostering a national market system and would be so harmful to the options market that CBOE could not accept it as a model for option market linkage. A mandatory routing switch is not a new idea. It has been raised several times in the past and each time the Commission has rejected it for very sound reasons. In fact, the Commission and securities market participants have discussed the effects of a mandatory routing switch in the context of equities since the adoption of the Securities Acts Amendments of 1975 ("1975 Amendments"). After the 1975 Amendments, the Commission was faced with a problem for equities similar to the current situation for options: the inability of the various segments of the securities industry to agree upon a common approach for linking the markets. In considering the various linkage proposals, the Commission rejected those that were based on a mandatory routing switch because such systems would in effect force all trading into a single mold. The Commission consistently has rejected mandatory routing proposals ever since then. For example, one initiative after the 1975 Amendments was the proposal of an electronic market system in which all orders would be entered into a computer-based system and would be executed automatically on the basis of strict time and price priority. In rejecting this proposal, the Commission explained that such a system "would have an impact upon existing market institutions which could properly be viewed as a fundamental change in the manner in which securities trading is now conducted, and it is difficult to foresee, and to provide against, the problems and difficulties which might arise." We believe the exact concern exists today for imposing a mandatory routing switch on options as it did for stocks in 1978. Forcing a firm to direct its orders based on a single dimension of multi-faceted competition inevitably will require pervasive regulation to hold all other dimensions of competition constant. Another example is the Commission's consideration in 1979 of a central limit order file, which the Commission proposed in the January 1978 Release. Although the NASD submitted a plan describing an electronic facility into which any qualified broker could enter limit orders for execution based upon time and price priority in the system, most other markets were concerned that absolute time priority would have significant deleterious effects on the exchange trading process and would eventually lead to the elimination of exchange trading floors by forcing all trading into a fully automated trading system. The Commission rejected this approach because of the possibility that introduction of a mandatory routing switch based on time priority could have a radical and potentially disruptive impact on the trading process, and decided that industry efforts instead should focus on the achievement of nation-wide protection of public limit orders based on the principle of price priority. In reaching this decision the Commission recognized that, if market makers in a particular market center have reasonable expectations of greater order flow by making markets that are consistently better in terms of price, depth, or ease of execution - not just price alone - they would be more likely to compete aggressively, thereby providing a better and more efficient market. The Division of Market Regulation recognized the drawbacks of trying to force a single system approach such as a mandatory routing switch on the equity markets in its Market 2000 Report in 1994. The Division did not recommend the adoption of such a system because it feared the consequence would be to stifle innovation and competition. The Market 2000 Report recommended that the Commission should be reluctant to impose a single design on the markets absent evidence of a significant market failure. We agree with this premise, and believe that no such failure exists in the options market. Any problems that might arise from the expansion of multiple trading are the consequence of an evolving business environment, and do not represent a significant market failure. These problems can be solved through a comprehensive linkage plan such as that espoused by CBOE, not a single market design like a mandatory routing switch that the Commission consistently has rejected. In the options markets context, the Commission has recognized that competition between exchanges can take place based on many factors, not just on the disseminated quote. For example, when approving the trading of options on Nasdaq stocks, the Commission stated that the benefits of inter-market competition include factors other than price alone, such as improvements in depth, liquidity and price continuity as well as more efficient execution, back office and clearing services. When the SEC last worked with the options exchanges on a linkage in 1992, four of the five options exchanges all except the Philadelphia Stock Exchange (PHLX) believed that a mandatory routing switch was not appropriate, for many of the same reasons we discuss in this letter. The four options exchanges filed a joint plan for the creation and operation of an Options Intermarket Communications Linkage, but the PHLX declined to sign the plan because the plan did not contain a mandatory routing switch. Unlike now, the Commission did not request the four exchanges to consider incorporating "time priority" into their plan. Indeed, the Commission did not approve any linkage plan, but instead directed the development of various systems enhancements and rule changes. In sum, the Commission has viewed an approach based on a mandatory routing switch as inconsistent with the statutory objectives of assuring fair competition and best execution, and contrary to the mandate of not forcing all trading into a particular mold. We see no reason why the Commission should deviate from its earlier positions. To more fully explain our rationale, in the remainder of this letter we explain why a mandatory routing switch remains undesirable. At the outset, CBOE respectfully disagrees with the characterizations in your letter of the possibility that a mandatory routing switch (i.e., time priority) would reward a market that posts the best quote first and could result in more aggressive competition and better prices. Quite the contrary, a mandatory routing switch would have the overall effect of reducing competition both between exchanges and within an exchange as it reduces incentives to compete other than on speed of quote change and would punish efficient markets that offer superior services and lower costs in addition to the best price. Simply put, a mandatory routing switch is unnecessary and anticompetitive. In understanding why a mandatory routing switch is so harmful, it is important to recognize all of the ways that exchanges compete for orders. One of the factors clearly is price, and at CBOE we believe that our Designated Primary Market Makers ("DPMs") and our other market makers (collectively, "market makers") quote very competitively. Such is a necessary, but not sufficient prerequisite, in successfully competing for order flow. There are numerous other ways that an exchange can compete for order flow. This competition has resulted in cheaper, faster, and fairer markets for all investors and professionals. A mandatory routing switch would significantly reduce an exchange's ability and incentive to compete in these other areas, leaving the quote as the primary means to compete. For example, some exchanges attempt to compete by providing greater depth of market and liquidity guarantees for orders. This enhances execution quality for all orders, even the difficult orders that may not be profitable at any moment in time to an exchange's market makers. Similarly, an exchange can compete by offering larger firm quotation and automatic execution guarantees. Order flow firms see these guarantees as beneficial to their customers' orders and consequently will route orders to the exchanges that offer these guarantees. In addition, the availability of member firm floor brokers and independent floor brokers to execute large or complex orders at an exchange makes that exchange a desirable place for firms to conduct business. Likewise, exchanges can compete by offering a lower cost of trading. There are numerous variables that contribute to the cost of effecting a transaction on an exchange, not merely its quote. These include the costs of maintaining a staff on an exchange floor, membership costs, floor brokerage costs, and transaction fees, among others. An exchange that is competitive in these areas will be attractive because it will offer a lower cost means of accessing liquidity. Lower costs not only benefit members but also their customers as well through lower brokerage commissions made possible by the lower overall cost of doing business on certain exchanges. Indeed, in response to the anticipated impact of screen-based trading in options, CBOE significantly reduced its exchange transaction fees, a competitive initiative by CBOE to which other options exchanges were forced to respond. Exchanges also compete by providing more efficient and reliable automation and order routing systems. The ability to handle orders quickly in an automated manner is an attractive feature in an environment where investors demand swift turnaround on orders and instant fill reports. The ability of firms to deliver their orders and receive fill reports quickly and reliably is even more valuable for options with heavy volume and during times of volume spikes. Consequently, CBOE has spent tens of millions of dollars to develop systems that benefit investors. CBOE will continue to devote significant resources to systems development because we know that systems performance is an important component of broker-dealers' decisions on where to send order flow. Exchanges compete for business in areas other than systems. Exchanges often attempt to provide customer service benefits to attract order flow. For example, a low frequency of errors and a quick resolution when errors occur make a firm more comfortable in routing customer orders to that exchange. Likewise, firms want to know that if a problem develops with an order, an exchange and its market makers will be responsive and fair. For this reason, the CBOE created the DPM system and recently expanded the DPM system to all equity options - because firms wanted a single point of accountability and responsibility. These competitive initiatives have reduced transaction costs, an important dimension of competition. Another area where exchanges compete is the quality of regulation, or self-regulation. The development of an automated audit trail, staffing of a large surveillance operation, and the maintenance of a member examination program all entail significant investments of personnel and capital, but it is money well spent because such efforts enhance the safety and fairness of an exchange and are consistent with firms' desire to do business on a well-regulated exchange. CBOE is the primary self-regulator in the options business. Surely the Commission will appreciate the investor benefits that result from an exchange's incentive to provide a well-regulated market. Finally, exchanges can compete by offering education and customer services. For example, CBOE has developed the Options Institute and the Options Toolbox for member and investor education about options. We have designed a comprehensive CBOE Internet Website that receives 50 million hits per month by more than 300,000 visitors. We provide these services because they enhance investor education, which we know is one of your highest priorities. Our member firms have told us that they place great value in the CBOE providing these services and want to see us continue to do so rather than have the services offered by an industry group. There clearly are other areas of competition or potential areas for new competition. All of these areas bring value to the options marketplace, and they all will bear weight in a firm's decision on where to send orders. The greatest beneficial impact of non-price competitive factors may be on small-sized customer orders. These orders are important to an exchange's viability as they provide a consistent stream of orders that are necessary to enable market makers to provide liquidity on an ongoing basis. Consequently, exchanges are especially motivated to offer superior services, enhanced systems and competitive costs in order to induce firms to route small-sized customer orders, even though these orders may account for a small portion of an exchange's overall contract volume. So long as an exchange guarantees the best price to customers, it should be allowed to attract orders by competing on all of these factors. What a mandatory routing system does, however, is to significantly reduce the importance of all of these factors, except price, so that the quote generated by an exchange at a particular moment in time is the only variable determining where an order is sent for execution. Every other element of intermarket competition would be rendered irrelevant instantly for those orders. Other markets and industries in the U.S. do not force competitors to compete on advertised price alone. Indeed, the irrationality of such a concept becomes evident when considering everyday commerce. For example, suppose the Department of Transportation ("DOT") imposed a mandatory routing switch on the airline business. Further suppose a consumer wanted his travel agent to purchase a roundtrip ticket between Chicago and Washington. A mandatory routing switch would leave the travel agent without any choice as to airline for the trip, as the agent would be forced to purchase the ticket from the airline that advertised the lowest price first. That airline might be a discount carrier that had a poor on-time arrival record, many customer complaints, a suspect safety record, and inconvenient departure times. The consumer may prefer to fly on the airline his agent usually books for him that has a 99% on-time arrival rate, high marks from customers, a stellar safety record, hourly departures, and a computer interface with the travel agent, especially if the airline were willing to match the discount carrier's price. A mandatory routing switch would prevent the agent from selecting the best airline by forcing the travel agent to place the consumer's flight plan into a system that would book the consumer on the airline that advertised the lowest price first. It is quite easy to predict the outcome of a mandatory routing switch requirement on the services offered by airlines. All airlines would quickly trim back their services and their commitment to quality and safety to the bare minimum because they could only compete on advertised price. Likewise, it is quite easy to predict the effect of a mandatory routing switch on the options exchanges. An exchange would have no incentive to provide deep and liquid markets, reliable systems, added services, comprehensive regulation or lower non-quote costs. Exchanges would compete solely on the speed of price changes. Costs would no longer be constrained by competitive considerations and services would degrade. There would be no incentive to provide competitive benefits if they were irrelevant to securing business. The options industry and investors would suffer greatly as a result. As described above, a mandatory routing switch would greatly reduce an exchange's incentive and ability to compete as an exchange. The natural consequence of this requirement would be the eventual destruction of the system of competing exchanges and the consolidation of order flow into a single, monolithic system of electronically competing orders. Established exchanges would disintegrate and market makers would compete solely by electronic quote generation in the series of options where order flow tends to concentrate, leaving series of options with less order flow (e.g. deep in-the-money series and LEAPS) without competitive quotes. Market makers ultimately would leave exchanges. A single processor would link the market makers and would result in a single point of failure with no incentive to innovate. Market makers would have no incentive to offer services beyond the bare minimum. This scenario should be very disquieting for you. Indeed, in your recent speech at Columbia University, you stated that discussion of our marketplace becoming an "execution utility" gives you pause, and that we should not allow any exchange structure to take hold that extinguishes the power of innovation. Arthur, we agree completely with you on these points. Even in the event that established exchanges did not disappear, the result would be equally discomforting. Exchanges that could not compete on the various service elements would be subsidized by an algorithmic system of order distribution. All exchanges would then degrade to the level of the "lowest frills" exchange, but, unless the Commission intervened, could charge high fees because order flow would be directed based on quote price only. Some exchanges would seek to game the system by imposing charges separate from the quote. Investors would end up paying more, but with inferior services. In order to prevent this, the Commission would have to intervene in exchange rate setting, but there would still be little reason for a new exchange to offer any of the value added services that exist today. The only entities that would consider registering as an exchange would be low-budget operations that competed solely on price and offered minimal services, systems reliability and regulation. Clearly, this is not how the Commission would want the options market to evolve. A mandatory routing switch would not only affect exchanges, but it also would undermine broker-dealers' relationships with their customers. A mandatory routing switch removes any professional judgment by broker-dealers in the handling of customer orders. The broker-dealer's role as agent for its customers always has been an integral part of the securities markets. CBOE supports the continued ability of firms' management to make critical judgments about the quality of exchanges' prices, technology, service, regulation, and other factors when directing order flow. Yet under a mandatory routing switch the routing decision would be transferred from the broker-dealer's judgment to an algorithmic system. Such a system would undercut a broker-dealer's best execution decision and interpose itself between the agent and its customer. Indeed, if a mandatory routing switch alone is used as the basis for order routing, what purpose do broker-dealers serve in trade execution? Why not permit investors to send orders directly to a central processor to route the order, with the receiving exchange and clearing house guaranteeing the execution, and bypass the broker-dealer? While this may sound farfetched, the implications of a mandatory routing switch upon brokers' fiduciary obligations could lead to this type of result. In addition, a mandatory routing switch would undercut a broker-dealer's ability to work orders to obtain a superior execution. A broker might determine that an order can receive a better execution if the order is worked by a broker on the exchange or might have a better opportunity for price improvement on a particular exchange other than on the exchange with time priority. Yet, the broker would not have that discretion with a mandatory routing switch because if it sent the order to a particular exchange, it would be rerouted in accordance with the time priority rule. Finally, a mandatory routing switch would sever the relationship between broker-dealers and exchanges. Firms would not be able to choose exemplary exchanges nor refrain from sending business to poor, inefficient or costly exchanges. Firms would be forced to transact with unresponsive and unreliable exchanges and market makers. Broker-dealers would have little incentive to remain members of an exchange let alone have any desire to participate in governance of an exchange. We know that most firms would not endorse a mandatory routing switch as a measure of fulfilling their responsibilities to customers. If provided an opportunity to present their views, they would renounce a mandatory routing switch. Indeed, one firm has already written you on the subject and expressed its objections to such a concept. We have solicited the views of many firms and we believe a strong consensus endorses a linkage approach that does not incorporate a mandatory routing switch. In attempting to portray the effects of a mandatory routing switch, we may sound alarmist, but that is because CBOE truly believes that a mandatory routing switch would irreparably damage the options market. With the conversion to decimals fast approaching, the exchanges should be devoting efforts to accommodate the conversion in a smooth manner while implementing an effective linkage. A radical redesign of market structure is the last thing the options exchanges need at this time as they try to address linkage, decimals, extended trading hours and expanded multiple trading. We understand that the SEC's interest in a mandatory routing switch stems largely from its concerns about the growth of payment for order flow and internalization practices in the options market. These practices have arisen very recently in the options market as some market makers have attempted to use inducements for order flow to attract business as a consequence of the recent expansion of multiple trading. Last month you expressed concern about the growth of payment for order flow practices for options. Rather than address the practice directly, however, the SEC appears to be contemplating whether to adopt a mandatory routing switch. Such a switch could reduce the practice of payment for order flow as market makers could not guarantee that purchased order flow would be routed to them. But, as noted above, this would be accomplished at a tremendous cost. CBOE agrees that payment for order flow and similar practices can be troublesome and may not always be in the best interests of the securities industry. We share your concern about the potential growth of the practice in the options industry. CBOE dislikes payment for order flow for many reasons, one of which is that it threatens to circumvent the clear service and technological advantage that CBOE has labored to achieve over 26 years. We adamantly believe, however, that the issue of payment for order flow does not warrant a radical restructuring of the options industry. Imposing a mandatory routing requirement to address payment for order flow would be like banning automobiles to end traffic congestion on the Washington beltway. Payment for order flow has existed for many years in the stock market. After years of debate about the practice, the Commission attempted to address the issue by adopting rules to require broker-dealers receiving payment for order flow in stocks to make certain disclosure to customers. At the same time, the Commission proposed rules to require the same disclosure in regard to options. The Commission never adopted the disclosure rules for options and has been silent on the issue in regard to options until very recently. Moreover, as you noted in a recent speech, the disclosure for stocks has been fine print boilerplate ever since the adoption of the disclosure rules. In light of the relatively relaxed SEC posture toward the practice for both stocks and options for the past several years, it seems anomalous to suddenly force the industry to solve the problem by adopting a system with as many negatives as a mandatory routing switch. Instead of imposing a government-mandated structure such as a mandatory routing switch, the SEC should address payment for order flow directly. CBOE understands that your staff is preparing a concept release on internalization practices to rekindle discussion on the issue. We strongly agree with the need to restart a public discussion on the issue. The problems raised by payment for order flow warrant renewed industry-wide examination of the practice. In the meantime, the Commission should be engaging in precisely the type of actions noted in your best execution speech: considering whether better disclosure would be helpful to customers and reemphasizing the need for firms' stringent adherence to their best execution obligations in their order routing decisions. We think that CBOE's linkage proposal would help firms in their efforts to seek best execution for options orders by preventing trade throughs, establishing intermarket price protection at the NBBO, and enhancing the firmness of options quotes, while preserving interexcha