Thank you for that kind introduction. And thank you for the invitation to speak at this meeting of the National Organization of Investment Professionals. These events are extremely helpful in promoting industry/regulator dialogue that is critical for the effective regulation of the securities markets. As usual, I must remind you that any views I express today are my own and not necessarily those of the Commission, any individual Commissioners, or my colleagues on the Commission staff.
As all of you are well aware, this is a busy time for those of us who work on market structure issues. Last fall, the Commission issued proposals relating to flash orders and dark liquidity. In January of this year, the Commission issued a concept release on equity market structure and a separate proposal on market access. Just a few days ago, the Commission issued a proposal relating to large trader reporting. That proposal is available on the Commission's website. In the near future, I anticipate that the staff will recommend to the Commission that it issue a proposal to establish a consolidated audit trail across all equities markets. It is conceivable that additional rulemaking proposals will flow out of the concept release. I will discuss each of these matters, but let me start by discussing some of the major themes of the concept release and the comments that the Commission has received thus far.
As the concept release notes, the Commission seeks market participants' views of the performance of the market structure as a whole, particularly for long-term investors and businesses seeking to raise capital. This should help provide context for particular concerns, as well as guide the nature of any regulatory response that may be appropriate to address those concerns. The concept release asks a series of questions about how the current market structure is performing and how market quality can be measured.
As of mid-April, the Commission had received approximately 140 comments on the dark liquidity proposal and the concept release taken together. Slightly under half of these comments were from individual investors. These commenters were deeply concerned about the current market structure. Some expressed scepticism about the very concept of dark pools. Others questioned the practices used by high frequency traders.
On the other hand, many representatives of both the buyside and the sellside generally expressed views that the current market structure, which supports multiple, competing trading venues, is beneficial overall. Some presented data supporting the view that market quality has improved over the last several years. For example, they state that quoted spreads have decreased, which may suggest that venues for undisplayed liquidity are not adversely affecting the displayed markets. For retail investors, a decrease in effective spreads is strong evidence of reduced transaction costs. Others point out that execution times have decreased substantially and that effective spreads have also fallen. Some observed that the equity markets performed well during the financial crisis in late 2008 and early 2009. They note that, although prices fell, price discovery remained active and holders of securities were able to trade at displayed prices.
Many, however, point out that there are still significant disincentives to quoting, and in particular to quoting in size. Because of price/time priority rules, the quickest actors are the first to execute — or are the first to get out of the way of adverse price movements. Thus, as these commenters argue, slower traders may face significant adverse selection problems if they submit non-marketable orders to a displayed market. Many commenters have observed that, when retail or institutional interest is quoted on a displayed market, other market participants can quote ahead of a displayed order by as little as a penny — which is permissible under Regulation NMS. These commenters take the view that there are significant disincentives for retail or institutional orders to be placed on displayed markets, which could diminish price discovery. This theory, if true, is troubling, since one of the primary aims of establishing the national market system is to promote order interaction and price discovery.
Others have pointed out that, even if spreads have narrowed, this does not necessarily mean that transaction costs overall are decreasing. Some would argue that size at the inside quote, at least for some stocks, is much smaller than previously, so larger orders still face significant costs to be fully executed. Others claim that large, electronically driven order flow can create intraday volatility that disadvantages retail and institutional investors. Under that logic, what an investor might gain from buying a stock when the spread is only a penny-wide is more than offset if the stock's price has been driven, whether intentionally or unintentionally by other forces, a nickel higher than its "true" price. The staff is working diligently to evaluate these comments.
On the subject of dark pools, some individual investors who commented are sceptical about dark liquidity of any sort. But institutions that invest on behalf of individuals are generally supportive. One industry group that represents the buyside stated that dark pools, "provide a mechanism for transactions to interact without displaying the full scale of a fund's trading interest, thereby lessening the cost of implementing trading ideas and mitigating the risk of information leakage." Buyside commenters have generally taken the view that dark pools are helpful in protecting their interest from opportunistic trading strategies.
With regard to high frequency trading generally, only a handful of commenters to date have specifically addressed the subject in depth. The staff is aware, however, that many people are studying this subject and an intense discussion of high frequency trading is going on in the industry press. A few recent studies have claimed that high frequency trading could hurt best execution, and could raise overall transaction costs for institutional investors. The staff looks forward to hearing the thoughts of those who are studying the matter closely. We are eager to further consider the findings and methodology of these studies, and hear any suggestions to address any potential regulatory concerns. Of course, the Commission has not yet reached any final views on high frequency trading or what, if any, regulatory actions might be necessary to address any issues raised by high frequency trading.
One of the issues frequently noted by commenters is that there is no obvious way of defining "high frequency trading." Trading frequently is not inherently bad, nor is using the fastest and most robust computer systems to process publicly available information to make trading decisions. We are interested in understanding whether — and, if so, how — high frequency trading might exploit certain market inefficiencies or regulatory gaps in a way that could disadvantage other investors.
One common observation relates to trading increments. As many have observed, the steep fall in stock prices beginning in the fall of 2008 has left a sizeable number of stocks that trade in the single digits. A penny trading increment becomes a greater percentage of the size of the trade as the stock price goes lower, so effective spreads of many sub-$10 stocks are comparatively high. Trading volume in many low-priced stocks has increased even as the stock price has fallen, which may increase the profitability of trading these stocks due to spread and rebate capture. One suggestion that has been floated within the industry is to lower the trading increment for certain stocks — say, using tenths of a penny for stocks trading between $1 and $10. Although the Commission has not made any proposal on this subject yet, and no market participants have put a formal proposal before the Commission, this is a subject that the staff will continue to monitor and study.
Another area where the Commission is working to address specific market practices that relate to high frequency trading is flash orders, the subject of a proposed rulemaking issued by the Commission last fall. Rule 602 of Regulation NMS generally requires exchanges to include their best-priced quotations in the public consolidated quotation data. The rule contains an exception, however, for quotations that are withdrawn if not executed immediately. This exception was originally adopted in 1978 for what then were considered the "ephemeral" quotations of traders on a manual trading floor. The Commission is concerned that the Rule 602 exception may no longer be necessary or appropriate in today's highly automated trading environment.
The flashing of order information could lead to a two-tiered market in which the public does not have access, through the consolidated quotation data streams, to information about the best available prices for U.S.-listed securities that is available to some market participants through proprietary data feeds. In addition, flash orders may significantly detract from incentives for market participants to display their trading interest publicly, though flash orders do offer potential benefits to certain types of market participants.
Over 90 commenters offered their views on the flash order proposal. Most commenters support elimination of flash orders in the equity markets and were in general agreement with the Commission's reasons for doing so. Some commenters, however, generally disagreed and others raised particular issues with prohibiting flash orders in the options markets. They cited differing market structure practices in the options markets, which resulted in lower cost executions for customers using flash orders. For example, access fees in the equities markets are capped, while no such cap currently exists in the options markets. Thus, according to the commenters, the benefits of using a flash order process in the options markets may be greater than in the equities markets. (I note, however, that the Commission just last week issued a rulemaking proposal to establish access fee caps in the options markets.) The staff of Trading and Markets will continue to analyze this and other issues raised by commenters as we work to prepare a recommendation for Commission consideration.
The dark liquidity proposal, as a notice of proposed rulemaking, was another effort by the Commission to address specific market structure issues that it has already identified — namely, practices that could allow unfair access to information about the best prices for equity securities and thus warrant greater transparency. One proposed amendment would require actionable IOIs to be treated like quotes. Another proposal would lower from 5% to 0.25% the trading volume threshold at which ATSs must display and provide execution access to best-priced orders. Taken together, these two proposals are designed to largely eliminate the potential unfairness in access to information created by actionable IOIs, where only selected market participants know about the best-priced orders.
Most commenters agreed that actionable IOIs should be treated as quotes. However, many commenters argued that more clarity was necessary to define what is an actionable IOI. Some commenters feared that examination and enforcement risk would be significant if the market cannot understand when it is permissible to send an IOI to selected parties. Still others asked whether it would be permissible for a smart-order router affiliated with an ATS to send IOIs that direct the recipients to send orders to the affiliated ATS. These are all issues that the staff will take into consideration as it prepares a final recommendation and drafts an adopting release for the Commission's consideration.
Regarding the proposal to lower the display threshold for ATSs from 5% to 0.25%, commenters were generally supportive. Some argued that the threshold should be only 1%, consistent with the threshold applicable to OTC market makers, or some other number less than 5%. However, others proposed eliminating it entirely. Some objected to the reduction to 0.25% but didn't present an alternative.
More issues were raised about the likely outcome of significantly lowering the display threshold for ATSs. Many doubted that this would lead to more quotes entering the displayed markets. They argued instead that interest that is currently IOI'ed would likely go completely dark, and result in more IOCs — in other words, venues that hold completely dark orders would be forced to blind ping other trading venues to find contraside interest. This could, they argue, risk even greater information leakage than could be caused by IOIs, as well as cause message traffic issues. However, some have raised the possibility that trading venues can impose fees based on a high level of cancellations to executions to dampen message traffic.
Each of the prongs of the dark pool proposal includes exclusion for large trading interest. Thus, the proposal would exclude from any quoting obligations any actionable IOIs for $200,000 or more that are communicated only to those who are reasonably believed to represent current contra-side trading interest of equally large size. Most commenters registered strong support for a size exception. However, many argued that a single threshold of $200,000 was too large, especially for small- and mid-cap stocks. Suggestions for improving the definition of the size exception include using a defined number of shares or a percentage of the security's ADV as alternate ways of qualifying for the exception. Some argued that the size exclusion should encompass small child orders that are part of a large parent order that qualifies for the exception.
A third prong of the dark liquidity proposal is to require real-time reporting by ATSs, including dark pools, of their identity as part of the consolidated trade data. Currently, ATSs must report their trades to the tape, but the tape print does not reveal which ATS executed the trade or indeed whether the trade occurred on an ATS at all. The Commission's proposal was designed to help investors identify sources of liquidity in particular stocks in real time, and to promote availability of reliable ATS volume information. However, many commenters, particularly on the buyside, opposed real-time attribution of ATS prints. These commenters generally stated that any marginal benefits to increased transparency were outweighed by the likelihood that institutional trading interest could be gamed by revealing in real time the venue where the trade was executed. There was, however, some support for end-of-day disclosure, although there are differing views as to whether it should be on a trade-by-trade or an aggregate basis. The staff is mindful of these concerns and suggestions as it works to develop a final recommendation for Commission consideration.
I'd now like to spend a few minutes talking some issues that overlap in the dark liquidity proposal and the concept release.
The fair access standard of Regulation ATS was not the subject of any amendment made in the dark liquidity proposal, although the Commission requested comment on fair access in the concept release. To date, no commenters have argued that retail or institutional trading interest suffers an inability to gain direct access to ATSs. As a general matter, dark pools welcome this kind of trading interest, and indeed some dark pools attempt to protect institutional trading interest by raising access barriers to the sell-side or certain hedge funds. A handful of commenters specifically lauded that the Commission did not propose any change to the 5% fair access threshold, arguing that it is appropriate to allow ATSs to discriminate against predatory trading interest that could disadvantage institutional users.
Various commenters have pointed out that institutional traders should have a better understanding of how their orders are managed and routed. The Commission has already stated in the dark liquidity proposing release that an ATS that markets itself as a dark pool yet sends IOIs to third parties regarding subscriber orders should adequately explain its use of IOIs to its subscribers. In addition, the Commission requested comment on whether — and, if so, how — disclosures under Rules 605 and 606 of Regulation NMS could be improved. Some commenters have offered their thoughts on that subject. Others have asked for more robust disclosures on Rule 10b-10 customer confirms. One of the things that could be achieved with such changes would be a better understanding of when and how customer orders are offered price improvement. Some commenters argue that "price improvement" is often meaningless. For example, a hundredth of a cent per-share improvement on a hundred-share order results in a customer "benefit" of a mere penny. Additional tools could be useful in helping investors understand whether price improvement being offered is significant or merely nominal. These are some key issues that the staff continues to study, and we hope to present some options to our Commission to address these points.
I'd now like to talk about some additional rulemaking proposals that may interest you. One proposal — relating to market access — was approved by the Commission for publication at the same time as the concept release. The proposal is designed to minimize the risks that broker-dealers incur when they access the market on behalf of themselves or their customers. Broker-dealers use a "special pass" known as their market participant identifier, or MPID, to electronically access an exchange or ATS and place an order for a customer. Broker-dealers are subject to the federal securities laws as well as the rules of the self-regulatory organizations that regulate their operation.
However, those safeguards do not apply to a non-broker-dealer customer who a broker-dealer provides with their MPID in order to individually gain access to an exchange or ATS. Under this arrangement known as "direct market access" or "sponsored access," the customer can sometimes place an order that flows directly into the markets without first passing through the broker-dealer's systems and without being pre-screened by the broker-dealer in any manner. This type of direct market access arrangement is known as "unfiltered" access or "naked" access. A recent report estimated that naked access accounts for 38% of the daily volume for equities traded in the U.S. markets.
Through sponsored access, especially "unfiltered" or "naked" sponsored access arrangements; there is the potential that financial, regulatory, and other risks associated with the placement of orders are not being appropriately managed. In particular, there is an increased likelihood that customers will enter erroneous orders as a result of computer malfunction or human error, fail to comply with various regulatory requirements, or breach a credit or capital limit.
The SEC's proposed rule would require broker-dealers to establish, document, and maintain a system of risk management controls and supervisory procedures reasonably designed to manage the financial, regulatory, and other risks related to its market access, including access on behalf of sponsored customers.
As of mid-April, the Commission had received over 50 comments on the market access proposal. The comments have been broadly supportive, particularly with respect to prohibiting naked access, although many commenters have raised specific issues. The staff of Trading and Markets continues to analyze these comments as it works to prepare a final recommendation for Commission consideration.
On Wednesday of last week, the Commission issued a proposal for large trader reporting, which is designed to provide the Commission with certain baseline information about major market participants, whether or not they are registered in any capacity with the Commission. Under the proposal, traders who engage in substantial levels of trading activity would be required to identify themselves to the Commission. A "large trader" would be generally defined as a firm or individual whose transactions in exchange-listed securities equal or exceed two million shares or $20 million during any calendar day, or 20 million shares or $200 million during any calendar month. Broker-dealers would be required to maintain transaction records for each large trader and report that information to the Commission upon request.
Chairman Schapiro has also asked the staff to work with the SROs to create a consolidated audit trail to enhance market surveillance. Certain mechanisms are already in place to coordinate surveillance among markets. For example, the Intermarket Surveillance Group — an industry organization created in 1983 — provides a framework for the sharing of information and the coordination of regulatory efforts among exchanges trading securities and related products to address potential intermarket manipulations and trading abuses. But given the development of the markets, there is general consensus that additional steps can and should be taken to improve intermarket surveillance. While I recognize such a step would require a substantial effort by the SROs and their members, a consolidated audit trail could be an invaluable regulatory tool to enhance the ability of regulators to detect illegal activity across multiple markets, and would greatly benefit investors and help to restore trust in the securities markets. With today's fast, electronic, and interconnected markets, there is a heightened need for regulators to have a robust consolidated audit trail and execution tracking system.
I want to briefly mention another subject which, though unrelated to equity market structure, is of importance to investors. The SRO rules governing analyst conflicts of interest currently are limited to equity research. However, conflicts of interest can arise in connection with debt research as well. For example, whenever a sales and trading department has a debt position, that department may have a heightened incentive to interact with the research department to pressure research personnel to publish favorable research so it can more easily reduce its debt position. Indeed, in 2008, a state regulator charged a prominent securities firm with wrongdoing, alleging that the firm's sales and trading department pressured its research department to publish favorable reports about auction rate securities in an effort to reduce its auction rate securities inventory. The staff has urged FINRA to consider expanding the current SRO research rules to cover debt securities, while recognizing that fixed income research differs in some respects from equity research and that those differences may suggest a different regulatory structure. I am pleased that FINRA staff has undertaken to address this issue, and I look forward to working with FINRA to address conflicts of interest in fixed income research. Investors are entitled to objective research, whether it's equity or debt research.
These are challenging times in the securities markets and complex issues we face. I cannot tell you with certainty what a perfect set of market structure rules looks like, or predict future rulemaking issuing from the Commission. But I am certain that the comments that market participants including those of you here today will inform the rulemaking process and benefit our securities markets and the investing public. Thank you for your time.