This document is available in three formats: this web page (for browsing content), PDF (comparable to original document formatting), and WordPerfect. To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.

Before the
Securities and Exchange Commission
Washington, D.C.


Release No. 34-42456, File No. 4-429

Proposed Option Market Linkage Plans by the American Stock Exchange, Chicago Board
Options Exchange, Pacific Exchange, and Philadelphia Stock Exchange

Comments of the United States
Department of Justice

I.   Introduction.

The Securities and Exchange Commission ("Commission" or "SEC") has requested public comment on the proposed Option Market Linkage Plans of the American Stock Exchange LLC ("AMEX"), the Chicago Board Options Exchange, Inc. ("CBOE"), the Pacific Exchange, Inc. ("PCX"), and the Philadelphia Stock Exchange, Inc. ("PHLX"). The Commission published this request in the Federal Register on March 2, 2000.

As the executive branch agency primarily responsible for promoting competition, consistent with other statutory goals, the United States Department of Justice ("Department") is pleased to have an opportunity to comment on the proposed option market linkage plans. The Department has conducted a number of investigations over the past decade into allegations of collusion in securities markets. As a result of these investigations, we have gained significant knowledge about promoting and maintaining competition in these markets. Our comments focus on the likely effects of the proposed plans on competition for customer orders. The linkage proposals grew out of a Commission Order directing the options exchanges to develop a linkage plan because the Commission had determined that "a linkage among options markets will benefit investors by increasing competition among markets (and market participants) to provide the best execution of customer orders." (Order Directing Options Exchanges to Submit an Inter-market Linkage Plan, Release No. 34-42029, October 19, 1999, at 5.)

II.  Summary of the Department's Comments

The Department strongly supports the efforts of the Commission to establish an inter-market linkage plan for multiply-traded options. We are in general agreement with the Commission's conclusion that "the options markets have developed sufficiently to make market integration not only possible but also critical to promoting vigorous competition among the options exchanges." (Order Directing Options Exchanges to Submit an Inter-market Linkage Plan, Release No. 34-42029, October 19, 1999, at 4.) Inter-market linkage is essential to increasing competition among options markets and to ensuring that customer orders are executed at the best available price. As a result, we believe the Commission's efforts will be of lasting benefit to the investing public.

In developing a system that promotes competition and protects investors, the SEC should consider whether all orders should be handled in the same fashion. The Department believes that a system providing price-time priority for small non-contingent orders would best achieve a competitive and efficient outcome for options market linkage. The Department believes that price is the dominant factor defining best execution for small orders, and that price-time priority would best stimulate price competition for these orders.

However, the Department is wary of mandating a system that provides for the same routing rules for all orders, regardless of size, specific contingencies, or the informed preferences of particular customers. Forcing a large order to go to a specific exchange based solely on price denies that order the opportunity to seek a market that has greater depth of capital. For a large order, depth of capital may be the most important competitive consideration. There may also be issues of trust or order-handling sophistication that might cause certain customers to use a particular broker on the floor of a specific exchange. These are legitimate competitive considerations, and the Department believes that customers who place a high value on these factors should not lose control of their orders based on a rigid rule of thumb that takes into account only the factor of price.

For these reasons, the Department believes that the best routing rules would have small, non-contingent orders executed on the best current market based on inter-market price-time priority. Larger orders would not be subject to a requirement that they be executed based on price-time priority. These customers could insist on cross-market price "matching," (i.e., that part of their order be executed at the best current price) for some portion of their orders, or not.

III.  The Proposed Plans

Three proposed plans have been submitted to the SEC: a joint plan from the AMEX and CBOE, a PCX plan, and a PHLX plan. They rely on a common framework, but differ with respect to important order routing rules.

CBOE and AMEX propose routing rules that would treat all orders in essentially the same manner. Under their proposal the original exchange receiving an order controls the order, regardless of size. If the original exchange does not have the best price, it may match the best price of any other exchange and execute the order, including the right to "step up" from an inferior posted price. Alternatively, it may choose not to and instead send an order to the exchange that has the better price.

All or part of an order may be sent to another exchange. If the original exchange sends an entire order, the second exchange must execute it at the posted price of the second exchange, up to the amount at which that exchange's posted prices are normally "firm." The second exchange has the right, for 15 seconds, to decide whether it will execute any additional amount of the order. If not, the second exchange must move its quote away, and the order (presumably) returns to the original exchange for completion. If the original exchange sends only part of an order, the second exchange can receive a second order by staying at its initial quote for more than 15 seconds after it executes the first order. If the second exchange moves its quote away within 15 seconds, the order (again presumably) returns to the original exchange for completion.

The system CBOE and AMEX propose will likely result in automatic price matching for all small orders through the receiving exchange's automatic execution system. Individual market makers who do not want to keep the small order at the "matching" price can route a "principal as agent" order to another exchange with a better price. In all likelihood, large orders will be considered individually. The ability to send large orders for execution at the other exchange's small order execution price will result in some traffic between exchanges for these orders.

The PHLX plan would also treat all orders in a similar fashion and would require execution of all orders based on price-time priority. Under this proposal, the exchange originally receiving an order would be obligated to route an order to another exchange displaying a better price. If several exchanges are at the best available price, the order must be routed to the exchange that first posted the best price. Under this proposal, the receiving exchange does not have the option of matching another exchange in order to execute an order. The plan would apply to large orders which today are usually handled by individually chosen floor brokers, as well as small orders.

The PCX plan would apply (during an initial phase-in period) only to orders up to 20 contracts.(1) The basic order handling rule under this proposal is price-time priority of a fashion. An original receiving exchange can keep an order, if it is not the best price in terms of price-time priority, only if it does as follows: (1) if it is at the best price, but not first in time, it can execute the order without regard to any other exchange if it improves its quoted price upon receipt; (2) if it is not at the best price, it can step up to the best price to execute the customer order, but only if it also sends an order equal in size to the exchange that had price-time priority before the order arrived.

IV.  The Department's Comments on the Competitive Implications of a Linkage Plan

The Department recognizes that price is not the sole criterion for measuring the quality of a trade execution. Depending on the customer, the characteristics of execution quality can include price, size, speed, accuracy, contra-party credit risk, and a range of administrative functions, including the quality of self-regulation and integrity that an exchange provides.

In considering the competitive implications of the proposed plans, the Department's primary concern is to preserve competition on price, size, speed of execution, and accuracy of execution.(2) The Department believes that it would be difficult to construct a system that takes into account investors' differing preferences for price, size and other criteria in a single, efficient, linkage system. We resolve this problem by proposing to limit the proposed rules for mandatory routing of orders to small orders. We explain the advantages of this approach, and the disadvantages of the other approach, in the next section.

The Treatment of Size in the Proposed Plans

Our principal concern with these proposals relates to an unspecified assumption in all of the plans, except to that of the PCX, that any linkage system should aspire to handle all customer orders -- regardless of size -- in an essentially similar fashion. However, the various factors that affect competition for customer orders can vary dramatically in importance depending on the size of an order, as well as on other features of an order for which size is often a good proxy (e.g. information, technical sophistication, and hedging advantages). In addition, it is clear that exchanges and market makers today make significant distinctions between large and small orders.

From the viewpoint of market makers, small orders justify little if any individual attention. On a trade by trade basis, small orders impose only a limited amount of financial risk. In addition, these orders provide essentially no market information in and of themselves. As a result, market makers are almost inevitably going to treat such orders as a category of trade for which they will post a universal public price on a take it or leave it basis.(3)

Moreover, all of the U.S. options exchanges already offer automatic computerized trade executions, at the current public quotes, for any customer market order that is below a specified size. It is the Department's understanding that this form of functionality -- automatic execution at a market's prevailing public quotes, for customer market orders below a certain size -- has become an almost universal feature of securities markets all over the world, including the U.S. equity markets. Such universal acceptance suggests that automatic execution satisfies the needs of both customers and market makers.

Larger orders receive different treatment in almost all markets. Once the size of an order exceeds the automatic execution threshold, it usually requires and receives individual consideration. A market maker may vary his price for such an order depending on the actual size of the order, the perceived risk that the contra-party may have an information advantage, a prior pattern of business, or any number of other factors.

In addition, distinctions between large and small orders can be found in the ways broker-dealers achieve their "best execution" obligations for their customers. In practice, almost all broker-dealers satisfy that obligation without individual bargaining by getting the best posted price available, so long as the order is small enough to fit within a "firm quote" obligation. Thus, if a system requires a participant to be good at his or her posted quote for some minimum size, a broker achieves "best execution" for any order at or below that size by getting an execution at the best posted quote.

For orders above the "firm quote" size, "best execution" is a much more complicated concept.

In many markets, such orders are subject to one-on-one negotiation. The Department is aware of no market in which there is a perceived fixed price that a market maker must honor for an order over the pre-set minimum size.

Stimulating Price Competition for Small Orders

As explained above, individual small orders usually do not merit individual attention. This does not mean that market makers or exchanges will not compete for this type of order. But it does mean that market makers are generally not going to adjust their market offerings in response to individual small orders.

This conclusion means that the only way to achieve competitive prices for small orders is to make sure that there is price competition for this category of order. The Department believes that price-time priority for small orders achieves this desired result, and that price matching could actually defeat this objective. Because these small orders are usually the least risky category of trade, inter-exchange price competition should generate better prices for these orders than for any other category. The Department also believes that including large orders within the same routing system could create pressure for exchanges to widen spreads for small orders, due to the increased risk imposed by the obligation to handle large orders (or a portion of such orders) at the same small order price.

The potential competitive benefits of price-time priority are apparent. As long as there are no qualitative differences in the item to be bought or sold, and no qualitative differences in the credit risk associated with any particular buyer or seller, price is the deciding characteristic of the quality of a transaction. Accordingly, any rule that dictates the routing of small orders should be based primarily on price. By using a "first in time" rule of priority, an exchange knows that it has to bid or offer a better price than that which is currently available, or it will not get the trade. The effect of the rule is to stimulate aggressive price competition.

If, instead, the routing plan permits matching, price competition would be dampened because the rewards that accrue from aggressive price setting are significantly reduced. An exchange need not fear losing a customer to a rival exchange if it does not have the best price. The exchange can always match the best price to keep the order. In addition, an exchange that does price aggressively may not attract any meaningful order flow if other exchanges decide to match. In volatile environments where rapid movements in the underlying stocks affect all participants' willingness to trade, aggressive price setting has risks. When matching is allowed in such settings, the risks are increased because a rival exchange has the option to match or send an order to the exchange with the best quote, and may send the order only when it is advantageous to it to do so. In such cases, the aggressive price setter will end up largely with trades that it regrets making. Market participants will recognize this possibility and price less aggressively. In contrast, if the aggressive price setter is rewarded with a large number of, or all, trades, then the rewards to aggressive price setting are increased and price competition should be more robust.

Consequently, the Department strongly favors a routing system that provides that small orders are routed immediately to the exchange that is first in price-time priority. Small orders in this context may be defined as any order for twenty (20) contracts or less. This threshold is emerging as the prevailing, minimum size guarantee for orders that may be automatically executed on exchange small order systems. Therefore, twenty (20) contracts would appear to be a reasonable threshold to set for price-time protection. We recognize, however, that size guarantees vary across systems and option classes. In some cases, orders as large as fifty (50) contracts may be routed through such systems. In our view, the threshold could be reasonably raised to a larger number of contracts or vary with the option class, although it may be desirable to have a common threshold across all exchanges trading that option class.

The Department believes that an exchange that improves price first should be given priority for all orders at that price as a reward for accepting the risk of price improvement. The options markets are dynamic and improving price poses risks and rewards. Because of the risks involved, an exchange should receive significant volume in order to create large enough incentives for price improvement. An exchange establishing a new, best price faces significant risk that the market will move away from its position. If the price-time preference is limited to a set number of contracts or for a set period of time, the incentives to improve price are reduced. Consequently, subsequent price matching by other exchanges should not affect routing decisions. Small orders should be routed to another exchange only when the first moves its market to an inferior price or when another exchange improves price.

The last exchange off a better price should have price-time priority at the inferior price.

If the Commission adopts a routing system based on price-time priority, it needs to consider which exchange has priority at a particular price if an exchange is moving away from what was a better price. The Department believes that the exchange that was last to move away from a better price, by, for example, being the last to lower its bid in a falling market, should be given priority at the new price. In our view, the market that was the last to move to an inferior price should be rewarded for having supported the market at the better price. The other possibility, giving priority to the market that was the first to move to an inferior price, would provide markets with an incentive to be the first to move off a better price.

However, a routing system that gives priority to the last exchange to move to an inferior price could potentially be manipulated to an exchange's advantage. For example, an exchange could gain priority by momentarily moving to a better price and quickly moving again to an inferior one. In order to preclude manipulations of this kind, conditions of some kind must be imposed. In this regard, the Commission might require that an exchange gains price-time priority by being the last to move to an inferior price only if it filled an order while at the better price. Alternatively, the Commission could require that the exchange have posted a better price for some set period of time, before moving to an inferior price, in order to gain price-time priority.

V. Conclusion

The Department fully supports the Commission's efforts to foster greater integration in the options markets. We share its desire to develop a system that protects and advances the interest of public investors. In our view, a system that routes small orders based upon price-time priority principles will maximize the benefits to the public.

Joel I. Klein
Assistant Attorney General
Antitrust Division

John M. Nannes
Deputy Assistant Attorney General
Antitrust Division

Timothy F. Bresnahan, Ph.D.
Deputy Assistant Attorney General
Antitrust Division

Mary Jean Moltenbrey
Director Of Civil Non-Merger Enforcement
Antitrust Division

George A. Rozanski, Ph.D.
Economic Regulatory Section

Dated: April 5, 2000

Respectfully submitted,

Nancy M. Goodman
Computers & Finance Section

George S. Baranko
Computers & Finance Section

T. Scott Thompson, Ph.D.
Economic Litigation Section

Patrick J. Greenlee, Ph.D.
Competition Policy Section
U.S. Department of Justice
Antitrust Division
600 E. Street, N.W.
Washington, D.C. 20530
(202) 307-6200


1. It is not clear when or whether the plan would be extended to larger orders.

2. The current organization of the standardized options markets depends on the Options Clearing Corporation ("OCC") serving as the ultimate contra-party for all customers. This feature eliminates contra-party credit risk as an issue.

3. As a result, small orders are usually executed automatically, without manual intervention. The CBOE and AMEX proposals recognize this and provide that orders not larger than an exchange's firm customer quote size are routed through its automatic execution system.