U.S. Securities & Exchange Commission
SEC Seal
Home | Previous Page
U.S. Securities and Exchange Commission

Speech by SEC Staff:
Remarks before the STA 71st Annual Conference


Annette L. Nazareth

Director, Division Of Market Regulation
U.S. Securities and Exchange Commission

Boca Raton, FL
October 8, 2004

Thank you so much for the opportunity to participate once again in the STA's annual conference. It is a particularly opportune time for me to be here since, as I am sure you realize, the Commission is actively considering the input it has received concerning proposed Regulation NMS. The STA represents the interests of traders throughout our markets and thus I know you are keenly interested in the ultimate outcome of the rulemaking. As always, I would like to sincerely thank the STA for your constructive input into this process. Before I begin my talk, I will remind you that my remarks represent my own views and do not necessarily reflect the views of the Commission or my colleagues on the staff.

The Commission's proposed Regulation NMS could lead to targeted yet important adjustments to U.S. equity market structure. In Regulation NMS, the Commission has proposed rules designed to account for the dramatic changes our markets have undergone in the recent past - particularly with respect to developments in trading technology and the rapid rise of alternative trading venues. The specific components of Reg NMS address the inter-market trade-through rule and price protection, inter-market access and access fees, market data revenue and its allocation among markets, and sub-penny quoting. Since you have already had ample incentive and opportunity to review the components of proposed Reg NMS and have voiced your opinions through the notice and comment period, I will not review the specifics of the proposed rules here today. Rather, I would like to spend my time this morning providing you with the big picture, a view from 10,000 feet if you will, on how the different parts of Reg NMS have evolved and provide a brief overview of my impression of the critical areas in which the Commission has received significant public commentary.

As you all know, the inter-market trade-through rule was adopted in 1978 as part of the ITS Plan covering the inter-market trading of exchange listed securities. The rule requires that ITS participants avoid trading on one market at a price inferior to that being offered at the same time on a competing market. The rule was designed to foster inter-market competition within a unified National Market System, while protecting limit orders across markets and enhancing the ability of broker-dealers to satisfy their best execution obligations.

With the advent of sophisticated trading technology and the related increase in the speed of trading in our markets, the ITS trade-through rule has come under increasing strain. The ITS rule was designed with traditional exchange floor auction trading in mind and its procedures highlight the inherent differences in electronic trading and traditional floor-based auction trading. The ITS rule permits a 30-second response time to inbound indications of interest; but for electronic traders, 30 seconds is an eternity. I believe it has become clear that, given lightning quick modern trading systems, the two models cannot co-exist peacefully under the ITS rule. A new regulatory paradigm must be developed that accounts for their inherent differences in market models and preserves their inherent strengths.

As you might have guessed, the Commission has received many thoughtful and articulate comments in response to proposed Reg NMS. I would like to spend a few moments giving you a flavor of the comments we have received and the issues we are considering.

Comments received in response to Reg NMS were most distinctly split with respect to the proposed trade-through rule. Overall, the majority of commenters advocated the principle of price protection. They expressed quite different opinions, however, on how the Commission should arrive at that end. Commenters seemed to fall into one of three groups regarding the need for, and structure of, a trade-through rule: (1) those who supported trade-through protection for automated markets' quotes and opposed any opt-out exception; (2) those who supported a trade-through rule but advocated an opt-out exception; and (3) those who did not believe any trade-through rule is necessary to protect the best prices and promote the display of limit orders.

Institutional commenters were nearly unanimous (with a few notable exceptions) in their strong support for a trade-through rule without an opt-out exception. They stressed that limit orders are the building blocks of public price discovery and efficient markets, and that an opt-out is contrary to the protection of limit orders. The major floor-based exchanges, other regional exchanges, and two of the three associations representing individual investors agreed with the institutional investors on the need for a trade-through rule. These commenters emphasized that only automated quotes deserve trade-through protection, however, explaining that quotes that are not immediately executable are not firm and should not be protected. They also felt that an opt-out exception would be unnecessary if only automated quotes received trade-through protection. Commenters argued that an opt-out exception could hurt retail investors, and allow participants to bypass legitimate, automatically accessible limit orders. These commenters supported a strong trade-through rule in order to protect and thereby encourage the use of displayed limit orders, thus promoting transparency, price formation, and liquidity in the NMS.

A second group of commenters supported a trade-through rule, but specifically conditioned their support on inclusion of an opt-out exception. Some commenters thought that a trade-through rule would provide valuable protection to retail investors, but institutional investors should be allowed to opt-out. These commenters further argued that an opt-out exception is an important tool for investors with different investing strategies, for example, those that value speed and those that are working large blocks.

A last group of commenters did not believe that any trade-through rule was necessary. This group included nearly all of the electronic markets and electronic traders, as well as the STA and a number of market making firms, and two large institutional investors. These commenters generally said that if market centers provide immediate and automatic trade executions against their published quotes and hidden access fees are eliminated, price competition would become the most meaningful standard of execution quality and market forces would then ensure that customers would receive the best price.

Commenters were split on whether a trade-through rule should be extended to the Nasdaq market. Commenters who supported a uniform trade-through rule stated that, by affirming the principle of price priority, such a rule would encourage the display of limit orders in all markets which, in turn, would improve price discovery and contribute to increased liquidity. Further, these commenters thought affirming the principle of price priority in all markets would increase investor confidence in the markets. Finally, these commenters argued that a uniform trade-through rule would facilitate a broker-dealer's ability to achieve best execution.

Commenters that opposed an expansion of the trade-through rule to Nasdaq, including the STA, questioned the need for trade-through protection in this market. They believed that competitive forces alone had already achieved the objectives of the proposed trade-through rule.

Certainly the most debated element of the trade-through proposal is the opt-out exception. Those who opposed the opt-out exception expressed concerns that it would allow block traders to by-pass limit orders at a better price posted by other investors, thereby discouraging the placement of limit orders and harming individual investors. Thus, these commenters believed that the opt-out exception would undermine the Commission's goal of price protection. Some commenters also suggested that the opt-out would degrade one of the core strengths of U.S. equities markets, the aggregation of retail and institutional investors' orders competing on the best price.

Many commenters supported the opt-out exception to provide investors with flexibility to pursue their own trading strategies. They also argued that an opt-out exception would promote greater competition among markets and would produce a more workable rule.

There was wide support for the concept of allowing automated or "fast" markets to trade through non-automated or "slow" markets. Many commenters specifically expressed support for a "quote-by-quote" exception to the trade-through rule that would allow market participants to trade-through manual or "slow" quotes. Some of these commenters felt the fast/slow quote distinction would allow hybrid markets to thrive. Commenters differed, however, on whether the Commission should set specific time standards for what qualifies as "fast." Those who opposed the manual market and manual quote exceptions seemed to do so because they favored the much broader opt-out exception (with fewer restrictions on the use than were set out in the Commission's proposal).

All that being said, as you know, the trade-through proposal is not the only piece of this market structure puzzle. The NYSE's proposed enhancements to the exchange's Direct+ automatic execution system are also part of the puzzle (and a bit puzzling). The Direct+ proposal clearly is intended to satisfy any trade-through requirement that would incorporate the automated vs. manual quote approach, rather than the automated vs. manual market approach. The NYSE's stated intent is for the Direct+ enhancements to significantly broaden the parameters under which the exchange would offer automatic execution of its displayed quotes. The NYSE hopes to position itself for a "fast/slow quote" trade-through rule, if the Commission chooses that approach, by preparing to be in a "fast quote" mode the vast majority of the trading day for each of the securities it trades.

I cannot predict what action the Commission will take on the trade-through issue. I can tell you with absolute certainty what the decision-making process will entail. The Commission will thoughtfully consider whether the benefits of a trade-through rule justify the costs and burdens. The Commission will carefully consider whether a rule can be crafted that achieves its articulated goals in a workable and targeted manner, that does not impede efficient trading, that furthers the goals of best execution and price formation, and that enhances investor confidence in our markets. It will analyze economic data and the impact on efficiency, competition, and price formation. It will weigh all of the comments made in the numerous open hearings and the public comment process. Given the rigor of this process, I am confident that any Commission action will strengthen, not weaken, the marketplace overall.

While the trade-through component of Reg NMS has clearly generated the most attention, the other three proposals have also attracted significant attention. The subpenny proposal would prohibit market participants from accepting, ranking, or displaying orders, quotes, or indications of interest in a pricing increment finer than a penny, except for securities with a share price of below $1.00. The Commission's stated goal in proposing the sub-penny component of Reg NMS was to change the status quo in which superior sub-penny quotes may be available on alternative markets, but are not transparent to and may not be readily accessible to average investors. At the same time, the Commission articulated its belief that including those sub-penny quotes in the best publicly disseminated prices could harm investors and the markets. Among other things, the Commission noted its concern that sub-penny quoting could decrease further market depth, increase the incidence of market participants stepping ahead of standing limit orders for an economically insignificant amount, and make it more difficult for broker-dealers to meet regulatory obligations because of "flickering" quotes.

The overwhelming weight of comment on this proposal appears to be in favor of the Commission's approving it as proposed. I agree with the majority of commenters that the marginal benefits of a further reduction in the minimum pricing increment are not justified by the costs associated with sub-penny quoting. I found the STA comments on this part of the Reg NMS proposal particularly constructive. Specifically, I agree with the STA that sub-penny quoting generally causes harm to the markets and may be used to step-ahead of competing limit orders for an economically insignificant amount. I also appreciate STA's practical perspective that sub-penny quotations increase the incidence of flickering quotes, thus making it more difficult for broker-dealers to meet their regulatory obligations, and that quoting in sub-penny increments can exacerbate a number of the less desirable consequences of decimalization.

With respect to inter-market access, Reg NMS would modernize the terms of access to quotations and execution of orders in the NMS by, among other things, establishing standards for direct and indirect access to quotes in the National Market System. The Commission's proposal on access fees would also place a 1 mil access fee cap with an aggregated cap of 2 mils on quotes in the NMS. Overall, the majority of commenters seemed to support the "soft linkage" approach, but comments were decidedly split with respect to the Commission's proposal on access fees.

Many commenters, including the STA, argued that all non-subscriber access fees should be prohibited. Nearly all electronic markets and electronic traders opposed any limitation on access fees, arguing instead that competitive forces were sufficient to address them. Institutional investors, some securities firms, and Nasdaq seemed to support the de minimis fee proposal as a worthwhile compromise approach on an extremely difficult issue. The SROs did not want their transaction fees to be subject to any limitation, beyond the current review of proposed rule filings by the Commission.

Without revealing specifically where I would come out on this issue, I will tell you that I believe market forces alone will not discipline all outliers in all cases. Without a cap or a ban, market participants will continue to be confronted with the difficult decision of either periodically paying outlier access fee charges or potentially violating best execution obligations. Indeed, a strict trade-through rule would strengthen the hand of the outliers. Thus, it is my hope that the Commission will, in the near future, adopt the access component of the Reg NMS proposal, including some resolution of the fee issue.

Finally, on the topic of market data, the Commission sought to address distortions caused by market data rebates by proposing amendments to the plans for disseminating market information to the public. The proposal, among other things, would modify the formulas for allocating plan revenues to reward markets for more broadly-based contributions to public price discovery. Commenters generally believed that there are, in fact, serious problems with the current formulas and the practices that have developed under the current formulas. Thus, there would appear to be benefits to updating these formulas.

Many commenters asked the Commission to consider the broader issue of the extent to which the amount being charged generally by the market data networks is reasonably related to the cost of producing the data. These commenters argued that market data revenues were far too high to be reasonably related to the costs of producing market data. Moreover, they pointed to market data revenue rebate programs as proof that the SROs were overcharging for their data. They questioned how significant percentages of market data revenue could be given back to members that contributed to creating the data if the SROs' costs in producing it were remotely related to the cost being charged.

As a threshold matter, it is clear that the issues surrounding market data, including its cost and wide availability to investors, are critically important. Market data is essential to investors and other market participants not physically present in a trading market, enabling them to make informed decisions when to buy and sell. It provides the basis for investment and portfolio decisions. Consolidated market information has been an essential element in the success of the U.S. securities markets. In addition to providing transparency of buying and selling interest, consolidated data is the principal tool for addressing fragmentation of trading among many different market centers, and for facilitating the best execution of investor orders by their brokers. It is clear that the pricing of this essential data is important to each of you. But we also need to consider these market data expenses in a larger context. For instance, total SRO market data revenue represented a very small portion of the securities industry's total expenses for the year - less than 1/4th of one percent - in 1998. While I agree that market data costs should be carefully considered, it is fair to say that market data costs are a very small percentage of overall industry costs.

That having been said, market data plays an important role in the funding of our self-regulatory system. There are limited sources of funding available to SROs. The primary sources are regulatory fees, transaction fees, listing fees, and market data fees. In response to the onset of stiff inter-market competition for order flow, members, and listed issuers, SROs have engaged in cost cutting and fee reductions. Fee cuts directly benefit market participants and investors, generally, but such cuts have significantly compressed SRO margins. Although SRO revenues have shrunk, SROs must continue to play the role that Congress assigned to them - front line regulators of the securities industry. Since satisfying this responsibility requires sufficient funding, market data revenue has become even more important now to SROs' funding than it has been in the past. In 2003, market data revenue represented 21% of SROs' total revenues. They represented 16% of NYSE revenues and 24% of Nasdaq revenues in 2003. For one SRO, market data fees accounted for more than 80% of its total 2003 revenue.

The costs of running SROs must come from somewhere. Keep in mind that transactions fees, regulatory fees, and listing fees charged by the SROs are under increased pressure. Market data revenue is unique in that it provides the broadest source of SRO funding. The fees are paid by all users of market information, including, for example, futures markets participants and media outlets that otherwise would not contribute to the funding of the particular markets on whose information they rely.

The Commission has clearly heard the concerns of commenters on the levels of market data fees. In order to better address this issue, I believe that the Commission soon will consider rulemaking that would require markets to make their market data revenues more transparent through better accounting of the revenues received for market data and the expenses incurred in operating and regulating the SRO's market. This transparency should provide observers greater ability to evaluate the role of market data revenues in financing an SRO, and to compare these revenues to the expenses of operating and regulating their market. This information also could empower users to respond to market data fee changes on a more informed basis.

We also plan to recommend that the Commission address the broader issues concerning market data revenue and SRO revenue, generally, in an SRO structure concept release, rather than in the context of adopting any final rules arising out of proposed Reg NMS.

These actions do not eliminate the original allocation concerns addressed by the Reg NMS allocation formula of reducing distortions from market data fees and rewarding markets for more broadly-based contributions to public price discovery. As long as market data is consolidated, and the fees charged exceed the costs of consolidation, allocation issues arise. The Commission's proposal is intended to address these issues in a thoughtful way.

In closing, I will note my belief that the industry and the Commission are at a critical juncture at this time. The opportunity to make adjustments to the way equity securities are traded in our markets well into the next century is upon us. You should be assured that whatever form final Reg NMS rules take, it is my sincere belief that they will reflect the Commission's continued effort to ensure that our markets retain their position as the most liquid in the world, to protect limit orders across markets, to design rules that facilitate the fulfillment of best execution obligations by agents in our markets, and to ensure that rules provide sufficient flexibility to allow industry innovation. It is my hope that the Commission will consider Reg NMS before year-end and that you will remain engaged in this process through the final rulemaking. Thank you.


Modified: 10/19/2004