Speech by SEC Commissioner
Remarks before the American Enterprise Institute

by

Commissioner Paul S. Atkins

U.S. Securities and Exchange Commission

Washington, DC
May 7, 2003

Good Afternoon. I'd like to thank Peter Wallison for arranging this conference and inviting me to speak with you this morning. Before I begin, I must first note that the views I express today are my own and do not necessarily reflect those of the SEC as an institution or of the other Commissioners.

With the completion of most of the Sarbanes-Oxley initiatives, I have looked forward to getting up here to speak about other matters that we at the SEC now have before us. It seems as if I'm always called upon to talk about what we've done regarding Sarbanes Oxley, so it's a pleasant change of pace to talk about some of the other important issues before the SEC.

Within the time that has been allotted to me, I cannot even begin to flush out all of the pressing market structure issues needing the attention and limited resources of the Commission. And, let's be honest, could I really talk about all of the topics even if it were an all-day speech? Luckily for you, you have some distinguished panelists to delve into these issues.

Still, undaunted by the task that lies ahead, I'd like to focus on at least a few critically important market structure issues that I hope the Commission will take up and address in the upcoming months. I'd like to summarize very briefly how our markets have evolved since the 1970s and how technology and competition have shaped the U.S. financial markets and created many of complicated issues that we face today. I would also like to talk about the impact of technology on the broker's best execution obligations and related market structure issues involving rules such as the trade-through rule.

The main message that I want to leave with you today is that the Commission should maintain neutrality regarding market structures. We should not favor one over another, because the bottom line is that there will always be people who want something different from everyone else, for whatever business or personal reasons, and are willing to pay for it. Competition among markets will drive innovation and progress, so long as there are no artificial barriers to competition.

In 1975, Congress and the Commission brought monumental changes to the United States equity markets. These changes included an end to government-protected fixed commission rates, the consolidation of quotes and last sale reports, and a basic linking of the listed equity markets. This was the original "Big Bang." Congress stated that its purpose in including these changes was to increase competition within the listed markets and increase the transparency and efficiency of securities transactions.

Today, we see the fruit of those regulatory changes, combined with the technological advances of the past 20 years, in the decreased cost of executing a trade, the broad montage of information that is distributed by various data vendors, the increased number of execution venues for listed securities, and the progression in technology both on and off the floor of the exchanges. Since 1975, the average daily share volume of NYSE securities rose from an estimated 18.6 million shares to about 1.4 billion shares. That is an increase by a factor of 50. The growth in the Nasdaq average daily volume is even more remarkable from an estimated 5.5 million shares to about 1.5 billion shares. That is an increase by a factor of 250. The dollar volume on all exchanges and on the Nasdaq increased 11,305% from about $179 billion in 1975 to about $20.4 trillion in 2002.

Since competition was the underlying justification for, and the by-product of, the 1975 Amendments, competition should continue to drive our decisions to seek further efficiency in the 21^st century. The Commission has initiated several market structure changes over the years, including the Order Handling Rules, Regulation ATS, the Repeal of Off-Board Trading Rules, and Decimalization. At the core of each of these initiatives was one very constant theme - an increase in investor protections through the simple introduction of competition.

Consider some of the fundamental issues that the SEC and markets are facing today: Intermarket linkages (not as advanced as they could be in view of today's technology), Market Data, Payment for Order Flow, Best Execution, the Trade-Through Rule, and Access Fees. Don't worry - I'll not try to discuss each of these issues today. But, each of these issues is subject to government-mandated regulation that may or may not make sense in a changing technological and competitive environment. They should be analyzed from the perspective that government's role should not be to favor one market structure over another, but to enforce core principles across all markets.

Since 1975, innovation transformed the Over-the-Counter market into a highly automated marketplace that thrives on competition. The old Over-The-Counter market that existed in the 1970s matured into a fully electronic, highly integrated organized market unlike its predecessor. Significant innovation in the securities markets has resulted from private companies competitively responding to changes in the regulatory and economic environment. From Electronic Communication Networks (or ECNs), to smart order routers, to direct access technology -- this innovation has been introduced in a market that is free from any intermarket linkage facility.

For you economists in the audience, we could probably call this recent period of dramatic technological innovation a classic "Schumpeter boom": entrepreneurs creating significant innovations that result in dynamic economic growth and change. Joseph Schumpeter, an Austrian who taught for years at Harvard, was one of the premier economists of the 20^th Century. He was also the very picture of "modesty." For example, he liked to say that he had 3 goals in life: first, to be the best lover in Austria (he was a notorious philanderer); second, the best horseman in Europe; and third, the greatest economist in the world. He said of himself that he was disappointed that he did not become the best horseman in Europe, but that he was happy that he had achieved the other two!

But, Schumpeter had some essential insights that changed the view of economists to competition and the face of antitrust regulation. He pointed out that price competition is not everything - that quality competition and sales effort need to be considered by government, rather than just a focus on prices, or else government will wind up forbidding activity that spurs growth and innovation.

Today, the increased efficiency in the market as seen in tighter margins for market participants, coupled with lower volumes, is bringing forth many competitive issues including access fees, locked / crossed markets, and equal access - to name a few. I am afraid that the industry's ability to adapt to the changing environment may be hampered by the uncertainty of the course of regulation. The Commission needs to remove that uncertainty by rapidly addressing the competitive issues that are important to the industry in these difficult times so that firms can continue to innovate and grow to meet the needs of investors.

Although the 1975 amendments primarily centered on improving transparency and linking the markets for securities listed on the NYSE and the AMEX, the Commission should now concentrate on establishing principles that can be applied broadly to all trading venues regardless of whether the product traded is a listed security, a NASDAQ security, or an option. Establishing uniform standards across all markets to create true equal regulation will strengthen investor protection and bolster fiduciary obligations. These uniform standards regarding best execution, the trade-through rule, and payment for order flow, regardless of the form, should be market neutral as well as product neutral.

Uniform regulation allows competition between the different market structures without forcing them into a single mold. Some individuals argue that this can only be achieved if the SEC comes to a resolution as to the definition of an exchange. We should provide that answer so that entities may reach business decisions but basic investor protection principles should be applied in all markets to produce the same results.

Decimalization is one of the reasons for the great debate over best execution and the trade-through rule. Before decimalization, total shares available for purchase or sale were displayed at every 6¼ cents (one-sixteenth of a dollar). Now, the market displays the total shares available at every penny. This breaking out of shares from the aggregate of a sixteenth to every penny as a separate price point instantaneously decreased transparency and liquidity within the market.

Investors now must access numerous price levels to obtain the same liquidity that was previously available through one transaction. Investors responded by requesting additional book information to replace the liquidity and transparency represented in the formerly aggregated price points. Investors also are demanding immediate access to these new liquidity price points similar to that offered by ECNs.

The New York Stock Exchange perceived and reacted to investors' needs and created data products and Institutional Express to fill them. Examples of such innovation include OpenBook and the recently approved Liquidity Quote. These data products improve pretrade transparency and increase market efficiency. In fact, a recent research paper by Ekkehart Boehmer, Gideon Saar, and Lei Yu indicates that OpenBook increased liquidity on the specialist's book and lowered transaction costs.

Decimalization, increased transparency in the market, and new trading strategies altered the way many investors interact with the markets. The emergence of electronic brokers allowed day traders direct access to the markets and the power to route their orders to their market of choice.

Today, the use of direct electronic access is wide-spread among traders in the NASDAQ market and is breaking into the listed market. This technology allows investors to make a real-time, best execution determination in order to fulfill different trading strategies. We at the SEC should address our rules that prevent proprietary traders from being able to reasonably access the best market based on their OWN weighting of factors such as price, speed, certainty of execution, and liquidity.

Let me illustrate how an individual can prefer speed over price with a personal story. When my wife was seven months pregnant with our first child during an exceptionally hot Washington August, my wife asked for some butter pecan ice cream with sour pickles (ugh!). We had only non-fat chocolate yogurt and sweet cornichons in the refrigerator. The local deli five blocks away had the favored brand of butter pecan and sour pickles, and my favorite chocolate chip cookies to boot. But, their prices were notorious for a huge mark-up, reflecting convenience, cost of doing business, and so forth. The grocery store a mile and a half away with lower prices only carried the store's proprietary brand of ice cream - a non-starter for my wife! The huge food warehouse on the other side of town carried everything, but it would take me more than an hour to get there and back. With that kind of delay, I would run the risk of bumping up against the next craving ... pizza with anchovies. Luckily for me, the local deli was there to meet my needs. Could you imagine if the government prohibited the deli from selling me the ice cream because it was cheaper at the food warehouse across town? Now try to imagine me explaining the ice-cream trade-through rule to my pregnant wife.

The essential question for us, and remembering Schumpeter's work, is whether securities regulators, especially as a result of the 1975 amendments, have been too focused on price competition and, through that single-mindedness, impeded improvements in the securities markets that competition can bring?

Well, like my experience with ice cream, the ITS trade-through rule may in fact prevent individual investors and professional traders from obtaining an execution that meets their needs. The brokers handling these orders are unable to provide their customers with best execution because they cannot trade through better priced bids and offers for immediacy or better liquidity which they may prize more highly than price alone.

In response, market structure in listed securities could change in several ways. For example, some have suggested that access to all markets could be standardized, meaning speed of execution and cost of access (or, in other words, price controls), so that the price of execution is the only variable. Thus, brokers would not have to balance the needs of their customers and could again route an order to a market center for execution based upon the best available price. This is not the best choice, because it does not emphasise the customer's needs.

The concept of standardized access is difficult to achieve, if it is even possible, and would interfere with the current market structure where various types of market centers compete for order flow. The basis for an auction market is to provide a place for buyers and sellers to meet on the floor of the exchange to obtain the benefit of liquidity and possible price improvement. Human interaction and negotiation is the central reason for benefits of sending an order to the floor of an exchange.

Any standardization in access that affects individual market structures should result from the natural course of competition and not from a government fiat. Another alternative may be to provide individual investors and professionals with the ability to set their own best execution priorities, whether it is best price, speed, cost, or liquidity.

The trade-through rule also ensures that limit orders in the market receive an execution. I recognize that allowing individual investors and professionals to choose where to send their orders could mean that better priced orders may be bypassed for execution. However, implicit in a broker's fiduciary obligations to its customers is a broker's obligation to ensure the best protection of investor limit orders displayed in the market. Absent the current trade-through guarantee, brokers would have to pay closer attention to the execution performance of market centers where they send their order flow.

In fact, today brokers have the very tools that they need to evaluate the execution performance of the various market centers because they are required to publish their execution quality statistics.

Best execution principles and the competitive fight for limit orders should provide sufficient protection for those accessible customer orders. This is best evidenced by the Nasdaq marketplace, which does not have an institutional rule to protect limit orders. It does not appear that quote competition has decreased in that market.

One interesting aspect of the growth of the national market system is the development of NASDAQ from SOES, to SelectNet, SuperSOES, and finally the completely automated SuperMontage. It wasn't until recently that NASDAQ faced tough competition from other exchanges trading NASDAQ-listed securities. Yet, NASDAQ faces many of the linkage issues presented to the Commission thirty years ago. The competitive solutions developed in the NASDAQ market through private industry are admirable.

Why does the trade-through rule exist for the listed and options markets but does not exist for the trading of NASDAQ securities? I won't bore you with a history of the US market structure since 1975. Rather, as I stated earlier, a uniform principle should exist between these regulated markets. The Commission should determine whether a trade-through rule is appropriate and then require it in all markets or abolish the rule altogether and rely on a broker's best execution obligation. One major proposition for displacing the trade-through rule is that automated markets can operate without taking time to seek an execution from a manual market.

The Commission took this very argument into consideration in August 2002 when it implemented the 3-cent de minimis exemption for three active ETF securities. Commenters in the industry also have asserted that the ITS trade-through rule inhibits meaningful competition in the listed market. One of the fundamental purposes of the 1975 amendments after all was to strengthen the capital markets by removing artificial barriers to competition that impede efficiency and unnecessarily interfere with the national market, investor protection being the exception.

According to the five-year Status Report of the 1975 Amendments dated August 1980, the ITS volume over a thirteen-month period as a percent of composite volume averaged 3.58%. The Status Report also stated, "[I]n making its evaluation of whether ITS is in fact the cornerstone of a national market system, the Oversight Subcommittee trusts that the SEC will recognize the point at which ITS is operating at its peak capability and decide, based on its characteristics, whether it meets the needs of the `national market system'".

Recent data show that in March 2003, ITS volume represented approximately 3% of the daily volume. So, 23 years after that 5-year Status Report, the minimal market share of ITS actually dropped - I will let you draw your own conclusions as to how effective ITS actually is.

I have to say that I echo the Oversight Subcommittee's 18-year-old Status Report that, perhaps after twenty-three years, it is high time to evaluate whether ITS and the trade-through rule meet the needs of the national market system. We have coming up an ideal opportunity to undertake such a study: our nine month exemptive order for the 3-cent de minimis exemption to the trade-through rule will expire June 4th. This would be a perfect time to take a broad look at the subject.

This study should review the overall ITS system to determine whether the alleged competitive restrictions of the trade-through rule do indeed exist, and if so, whether the rule impedes the efficiency of the markets. The study should also review whether this regime adequately protects customer limit orders or whether, in theory, it should, but, in reality, it does not. Or, will regular market forces via arbitrage trading strategies provide accessible customer limit orders with adequate protection?

I firmly believe that the Commission should continue to facilitate the national market system by allowing competition to form the structure of the securities market. I must say that the securities industry is indeed competitive enough to continue to innovate and increase market efficiency. I know that the Division of Market Regulation will be providing recommendations to the Commission within a short time period regarding issues including Market Data, Payment for Order Flow, Best Execution, the Trade-Through Rule, and Access Fees. These might be some of the most intractable issues that we face in the market structure area. If I left out a topic, I am sure that you all will help to point it out!

As I stated earlier, my decision on any reform of the national market system will be guided by the principle that the Commission should remain neutral and not adopt rules that favor one market structure over another. I firmly believe that competition itself will determine the appropriate market structure as long as there are no artificial barriers to competition. Whereas the mentality in 1975 centered on how to break up the New York Stock Exchange's monopoly, the feeling today is how do we allow market participants to continue to innovate while protecting investors and serving their needs.

Again, I thank you for your obvious interest in and concern for this subject that you show by your presence today and by your attentiveness.