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

Speech by SEC Commissioner:
Remarks Before the Security Traders Association of Boston


Commissioner Paul S. Atkins

U.S. Securities and Exchange Commission

Boston, Massachusetts
September 13, 2007

Thank you, John (Gisea) for that kind introduction. It is an honor and privilege to be here today with you in Boston. Before I get started, I need to satisfy my own compliance people and tell you that the comments that I am about to make are my own and do not necessarily reflect official SEC policy or the opinions of my fellow Commissioners.

My three Red-Sox fan sons do not usually want to accompany me on business trips, but last night they were very jealous about my coming to Boston today. That is, until they found out that the Red Sox are playing New York at Fenway Park tomorrow and not today. They wondered if John might want to postpone this meeting until tomorrow! Anyway, all the best to you all as you head into the playoffs.

What an amazing time to be speaking to a group of securities traders! The markets have been a rollercoaster ride for almost a month now due to a series of events triggered by problems in the subprime mortgage market. Time will tell us if we are experiencing merely a correction or a market event comparable to what we experienced in the late 1990s with Asian currencies and the Russian debt markets. I will make no predictions except one: all of our jobs will continue to be interesting in the near future!

Before I move on to my remarks, I would like to take a moment to reflect on the sixth anniversary of the September 11, 2001 terrorist attacks. Six years ago, the country experienced its greatest trauma in generations — an event so horrific and shocking that many Americans wondered if we could ever recover. Although clearly secondary to the emotional anguish, the effect of September 11 on the financial and trading markets was immediate and profound. Six years later, though, despite the fallout from the recent credit crunch, the economy is in good shape and we can look back at several consecutive years of aggressive economic growth based on sound fundamentals.

This is a testament to the will and courage of the American people. And, the President certainly built a strong basis for recovery and growth through a solid monetary and tax policy. But, we would not have been able to dig out from the events of September 11 without the dedication and heroic efforts of a select few. We owe tremendous gratitude to all of the people in uniform, be they civilian or military, who in the six years since have made such enormous personal sacrifices to defend our freedom against those who would destroy us.

Aside from the recent market disturbances, there are major structural changes taking place here in the U.S. markets, as well as in the international markets, some of which I will talk more about in a minute. Significant business and regulatory changes are also afoot.

  • The NASD and NYSE Regulation recently have completed their merger.
  • The NYSE and Euronext have also merged, and will presumably now seek to capitalize on the potential efficiencies that the combination might offer.
  • The NYSE is back in the bond trading business.
  • NASDAQ appears to be developing a taste more for Swedish vodka than for English gin.
  • NASDAQ looks to be creating and/or buying an options market and has reinvigorated its PORTAL system for 144A securities.
  • Institutional trading firms like BATS, ITG, and GETCO represent a significant portion of the NYSE and NASDAQ trading volume.
  • And, the way some other people have been talking — all around us, in the back alleys and in the unlit corners, "dark pools" are multiplying! For the record, Alfred Hitchcock saw this coming. A 1963 episode of The Alfred Hitchcock Hour was entitled "The Dark Pool."

All of these changes, along with the recent subprime problems, are happening at a time of concern about the competitiveness of the U.S. capital markets. This concern has been manifested, in part, in three major reports on U.S. competitiveness. Each of these reports contains recommendations, including some aimed at improving SEC rules and procedures. A constant theme in the reports is that excessive, overlapping, and unnecessary regulation in the U.S. is a major reason for our loss of market share in the global capital markets.

Although the perspectives and findings of each report are unique, there is a common thread of very important SEC-related issues among them. Three of them stand out: fix the implementation of section 404 of the Sarbanes-Oxley Act — which I hope we now have done with the approval of Audit Standard 5 (although the jury is still out on that); use cost-benefit analyses in making rules; and work with the other financial regulators to bring transparency and predictability to the enforcement process. To tell you the truth, these recommendations are not radical — they are common sense measures that the SEC should undertake whether or not they affect U.S. competitiveness.

We at the SEC cannot and should not ignore these findings and recommendations. We must recognize and understand how the markets have evolved when we consider whether at least some aspects of our weighty regulatory precedent still make sense. Times change and we ought to review our rules to see if they still meet the purpose for which they were made in the first place. We need to ask ourselves the same question that Secretary Paulson has posed: "Have we struck the right balance between investor protection and market competitiveness — a balance that assures investors the system is sound and trustworthy, and also gives companies the flexibility to compete, innovate, and respond to changes in the global economy?"2 The reports can help us answer this question.

To that end, I am pleased to say that the SEC has already planned to take a hard look at two of the issues raised in the reports. Specifically, the SEC recently created an Advisory Committee on Improvements to Financial Reporting. The Committee, which is chaired by Bob Pozen, will focus on the following issues: (1) the current approach to setting financial accounting and reporting standards; (2) the current process of regulating compliance with those standards; (3) the delivery of and access by investors to financial information; (4) factors that may drive unnecessary complexity and reduce transparency; (5) the costs and benefits of existing accounting and reporting standards; and (6) any lessons that can be learned from the growing use of international accounting standards. That is certainly a full plate. I am very excited about the work the Committee will be undertaking, and I look forward to the recommendations that the Committee will make to simplify the financial reporting process.

Separately, Chairman Cox recently announced that the SEC will be organizing a formal roundtable to explore the topic of private securities litigation. This issue was discussed in each of the three competitiveness reports, and was uniformly noted to be a major problem affecting the competitiveness of the U.S. capital markets. I am greatly looking forward to this roundtable discussion, and I expect that we will hear an interesting and useful exchange of ideas. Private securities litigation can play an important role in helping to sort through financial fraud and even providing a deterrence against fraud, but a system that facilitates meritless litigation can impose tremendous costs on investors and deter innovation and productive economic activity.

I believe that the Commission is duty-bound to analyze, understand, and — if warranted — respond to these and all other report recommendations that pertain to us. Unfortunately, some nay-sayers reject not only the recommendations of these reports, but also the very need to examine our regulatory scheme. I welcome this debate, and I encourage all of you to pay attention and to be involved, because these issues will directly affect your profession and the shape and strength of the American capital markets in the years to come.

In the end, companies are rational — they expect benefits of the public capital markets to exceed the costs. If that is not the case, they will go elsewhere or raise capital in the deep and flexible US private markets. This was proven true in 2006, when the value of Rule 144A unregistered offerings in the U.S. for the first time exceeded the value of public offerings. Our job as regulators is to examine the costs that we impose on market participants through our regulations to make sure that those costs do not exceed the benefits.

I am pleased to say that there have been recent glimmers of hope when it comes to the SEC's taking an economic view of rulemaking. Perhaps the best example of this is the review undertaken by the SEC's Office of Economic Analysis (OEA) regarding the effectiveness of the short sale rules. For many years, academics had been questioning whether the SEC's short sale price test, or "tick test," was needed. The tick test, which has been on the books for seventy years, restricts the prices at which short sales may be executed.

Our economists conducted a pilot test on approximately 1,000 securities to investigate whether removing the price tests would make stocks more susceptible to patterns associated with downward manipulation. They did not find evidence to suggest that pilot stocks are more likely to be manipulated downward than other stocks. Based on the results of the pilot, the SEC proposed last December to eliminate the tick test across the board. I commend the ingenuity of the staff — both in OEA and the Division of Market Regulation — in formulating this approach, and I look forward to seeing similar empirical methods used in the future.

Other recent rulemakings by the SEC can be neatly described by a single, four letter word that features prominently in the nightmares of securities operations professionals — the word is, of course, "FAIL." Fail is an especially appropriate word to describe the substance of one of these rulemakings, and the process and theory of others. The first rulemaking involves the noun form of "fail," as in fail to deliver. I am referring to the recent amendments to Regulation SHO, which were a very positive development. I am happy to report that the amendments we adopted should help reduce the number of aged fails to deliver. These amendments eliminated the now-notorious "grandfather" exception from Reg. SHO and limited the scope of the options market maker exception.

The decision to amend Reg. SHO was based upon a thorough and extended review of empirical data on fails collected by the SEC staff after implementation of Reg. SHO in 2004. In a time when the SEC has come under intense scrutiny for shortcomings in our rulemaking processes, this thoughtful, coordinated approach was truly a breath of fresh air. It exemplifies the expectations of Congress when it placed on us the responsibility of considering whether our actions will "promote efficiency, competition, and capital formation."

Unfortunately, proactive regulatory thinking does not characterize all that the SEC does, and that is where the verb form of the word "fail" comes in — as in the SEC failed to do its job properly and was slapped down by the courts. These are mainly the regulatory missteps that the SEC made before the arrival of Chairman Cox. During that time, a majority of the Commission — when I found myself in the minority — promulgated several rules without adhering to Congress's requirements for our rulemakings — including the not trivial requirement that the SEC act within the authority granted it by Congress. Two of these rules, the hedge fund registration and mutual fund independent director rules, were challenged and defeated in federal court. A third, and the most relevant to this crowd — Regulation NMS, has so far successfully escaped court review, but that is the only way in which it has succeeded. In Reg. NMS, the SEC failed to establish a need for rulemaking, failed to conduct a meaningful cost/benefit analysis before promulgating the rule, and failed the U.S. markets by adopting the rule. Its implementation process has further been characterized by a failure to meet the rule's own timeline for effectiveness of its various components.

In a nutshell, Reg. NMS rules fall into four categories: (1) the trade-through rule (it is an understatement to say that the sweet-sounding appellation "Order Protection Rule" is a misnomer); (2) the access rule; (3) the sub-penny rule, and (4) the market data rules. Some of these rules are much more problematic than others.

In my view, Reg. NMS represents a massive regulatory intrusion into our secondary trading markets that was completely unwarranted, given the lack of evidence of market failure and the availability of substantially less-intrusive means to advance the purported goals. Reg. NMS has the potential to do significant harm to our markets by unduly interfering with the operation of competitive forces. Over the years, these forces have benefited investors immensely by reducing trading costs and increasing market efficiency. Whatever its justification, Reg. NMS is a carte blanche for unsupervised meddling by the SEC staff in the marketplace for years to come.

The most problematic component of Reg. NMS is the trade-through rule. Based on the trade-through studies I read three years ago during the rulemaking process, I did not see a need for trade-through "protection." Protection from what? The freedom to choose how you want your trade executed? Does the government know better than investors how trades should be executed? I don't think so. And, one need only look at the precedent for this sort of rule — the old SRO/ITS trade-through rule — to see that the only thing "protected" by the rule was an outdated market structure. And to make matters worse, the Commission dropped from the final rules a proposed opt-out provision that would have allowed market participants the ability to choose where to execute their trades and at what price.

Market participants did not need the "protection" offered by Reg. NMS's trade-through rule. As your sister group in New York — STANY — so aptly put it in their 2004 NMS comment letter: "there will be no need for a uniform trade-through rule if issues of connectivity, access, and automatic execution are adequately addressed."1 Importantly, the comment letter then stated that "if the [NBBO] in every market is immediately available to 'away markets' then STANY believes broker-dealers' best execution obligations would be sufficient to protect the interests of all investors and ensure that superior prices are sought." I completely agree with this position. And so I am sorry to report that now — despite the fact that competition has driven the manual markets to automate — we have a trade-through rule and at the same time no assurance of best execution compliance for orders sent to some of these automated markets.

My opposition to the adoption of Reg. NMS does not in any way make me less interested in the implementation process. Quite to the contrary, I have been fully engaged in the process to ensure that the rules are implemented in the least burdensome and most cost-effective manner possible and in a way that limits their potential harm to our markets' competitiveness. I am grateful that market participants have been providing crucial feedback to the SEC on Reg. NMS implementation. This feedback has been invaluable.

As most of you know, the original compliance date for the trade-through rule and market access rules — which together are the heart of NMS — was June 29, 2006, well over one year ago. As with other, less-important components of Reg. NMS, this deadline has been repeatedly and significantly delayed. The first extension, which the Commission approved in May 2006, replaced the 2006 deadline with a complicated phase-in schedule. I said last year that I doubted that the industry could meet even these more generous timeframes. So, it was no surprise this January when the New York Stock Exchange requested a further extension of the February 5 trading system deadline. The Commission approved a one-month extension of that deadline, even though the additional one month period did not afford the NYSE enough time to remedy its inability to route orders to the ADF and ISE! The Commission also extended the start date for industry compliance for 250 stocks from May 21 to July 9, and the start date for full industry compliance from July 9 to August 20. The October 8 deadline for completion of phased-in compliance with Rules 610 and 611 remains in place — at least for now!

Why all of the delays and extensions? The difficulty in the creation of new, electronic trading systems by the exchanges was undoubtedly one reason for the extensions. However, the truth of the matter is that the unduly complex and burdensome nature of the rules themselves made the extensions necessary. The implementation process already has been a rude awakening for many, and we have not finalized the phase-in for broker-dealers yet!

Technology is only one aspect of the implementation burden. The SEC staff, using the extremely broad authority that the then-majority of the Commission so casually delegated in Reg. NMS, has provided several exemptions to facilitate the implementation of NMS, including significant exemptions from Rule 611 — the trade-through rule. One such exemption is for "Qualified Contingent Trades." According to the SIA exemptive request, a contingent trade is "a multi-component trade involving orders for a security and a related derivative, or, in the alternative, orders for related securities, that are executed at or near the same time." These spread trades are hugely important to investors, and they are a substantial revenue source for broker-dealers. I am pleased that the SEC staff finally found the wisdom to exempt these trades from the trade-through rule. It would be a bizarre result indeed for market participants in a spread trade to have to seek out and satisfy "protected" bids or offers.

Another important exemption from the trade-through rule is the "print protection" exemption. According to the SIFMA exemptive request — this came after the BMA/SIA merger — print protection "is the mechanism through which broker-dealers may elect to execute a displayed order at a price that is better than a reported trade in the same security on a different market." Sounds logical. Why should a rule prohibit a broker-dealer from improving the price, right? Well, that is where one of the trade-through rule's unintended consequences comes into play. If a client's order, say it is a buy order, is posted but not at the top of the book, it is possible that a sell order could pass over the client's buy order and instead execute against a worse-priced bid. This happens when an Intermarket Sweep Order — itself an exception to the trade-through rule — is sent to take out the top-of-the-book quotes, missing better-priced quotes within the book.

It is only natural — and maybe even compelled by the duty of best execution — that a broker-dealer would want to fill its customer's order after it had been passed over for a worse price. It may seem unbelievable, but that would not have been possible without the print protection exemption, because such trades could have resulted in a trade-through violation. In approving this exemption, the SEC staff, again using delegated authority, noted that "[p]rint protection… can improve the execution of depth-of-book quotations and thereby promote price discovery." If you untwist the SEC staff's logic, the real message is that the trade-through rule can impair executions of depth of book quotations and thereby impair price discovery. Thus the need for an exemption.

There are other, important exemptions, such as the sub-penny and error correction exemptions. There are also the nine exceptions that are "built in" to the trade-through rule, such as the benchmark, ISO, and stopped-order exceptions. Each of these exceptions is further detailed and nuanced in sixty or so FAQs.

So now you have it. We now have had to resort to a slew of exemptions, exceptions, and FAQs to tailor an inappropriate, "one size fits all" trade-through rule to the needs of a dynamic marketplace. At bottom, however, these carve-outs, along with the implementation delays, are the clearest sign that NMS — the trade-through rule in particular — was unworkable as proposed. As we dissenters said at the time of adoption, we have no way of knowing what innovations will be stifled by this rule. Just look at how long and how hard the industry has had to work to get these changes approved.

So, I declare here, for the first time, that the inevitable has come to pass: The trade-through rule is mortally wounded. It has suffered a thousand cuts, but it is finally close to lifeless — limping along like an old man. The exceptions and exemptions have inexorably eaten the rule! The trade-through rule has been eviscerated by the carve-outs demanded by a dynamic market system, and all that is left is the still-too-burdensome, unneeded husk of a bad idea. The reality of the situation has not been lost on market participants — let's just say that I have heard this same thing from numerous anonymous sources. Now, mind you, some at the SEC do not agree with this assessment and are still searching for vital signs. I am not holding my breath for a formal surrender ceremony such as the one General MacArthur experienced in Tokyo Bay on the deck of the U.S.S. Missouri 65 years ago this month. The following quote from a recent New York Stock Exchange filing creating a "Do Not Ship" order type illustrates the charade that the trade-through rule has become:

The Exchange states that . . . orders that are routed away to other market center(s) in compliance with Reg. NMS may cause the market participant to incur multiple fees because the customer has to pay a separate fee each time the order is routed to other market center(s) during the course of its execution. The DNS order enables a market participant to control the costs associated with order execution by limiting the execution of the order in whole or in part, to the Exchange.2

What does this mean? New York filed this rule change because other exchanges had already implemented it. So, we are now at a point where a customer can effectively opt out of the NMS trade-through rule and can direct his execution to the exchange where he wants it to go for the price that he wants. We have come full circle after a huge detour and several hundreds of millions of dollars of implementation costs incurred.

It took the SEC only three years and countless hours of industry time to circle back — in a Rube Goldberg way — to the basic premises of the opt-out rule that we originally proposed in 2004. I continue to believe that an opt-out provision is a much-needed addition to Reg. NMS. It would certainly be a more straight-forward approach. The problem is that, taken together, the exemptions and exceptions lack the simplicity and breadth of the opt-out provision. Perhaps it is time for the Commission to consider returning to first principles.

I will continue to watch carefully the implementation of Reg. NMS and hope that its consequences turn out to be a non-event — other than the thousands of hours and multiple millions of dollars spent in an effort by market participants to implement it. The trouble is that the real consequences for our markets will not be known for years, after the damage has been done and after trading patterns have shifted in unpredictable ways and perhaps to other venues.

You have been a very patient audience, and I appreciate your attention. I welcome your active involvement in the issues I discussed today, and all of our issues. My phone and office are always open to you. Please call or stop by if you have any comments or concerns. Thank you again for your time and attention.



Modified: 09/14/2007