Addressing Market Instability through Informed and Smart Regulation

As many of you know, I am now in my second term as an SEC Commissioner and this is my fifth time participating at SEC Speaks. During that time, I have served with three different SEC Chairmen, and a fourth is now in the works. It has been, and continues to be, a great privilege to serve at a time during which the SEC’s role as the capital markets regulator has never been more important. However, I must admit being frustrated that we haven’t done more to protect investors.

Clearly, my tenure as a Commissioner has been dramatically impacted by the financial crisis and the pressing need to address the many failings that were brought to light by that crisis. Throughout my tenure, I have worked to be a strong advocate for fulfilling the Commission’s mission to protect investors, facilitate capital formation, and promote a fair and orderly market. To that end, I want to talk to you today about the need to protect investors through robust and effective market oversight.

I am growing increasingly concerned about the stability of our market structure as we lurch from one crisis to another, be it the flash crash or the Knight trading fiasco. Today, I plan to focus on the dangers that investors face from a trading market structure that has shown too many signs of weakness and instability.

The Evolution of Our Markets

Over the past two decades the securities markets have experienced significant changes, evolving from a collection of a few, mostly manual markets, to a large number of high-tech trading centers. The new trading systems that have been developed by exchanges and alternative trading systems (“ATSs”) rely almost exclusively on fully-electronic, automated technology to execute trades. The speed and capacity of automated systems has contributed to dramatic increases in trading volume. For example, the average daily volume on the New York Stock Exchange in 1990 was 162 million shares. Today, the average daily volume on the New York Stock Exchange is approximately 2.6 billion shares — a 16 fold increase.

In addition to these increases in speed and volume, there has been an increase in the number of trading venues. No longer are trading activities controlled by a few trading venues, as it was two decades ago. Today, equity trading occurs across 13 national securities exchanges and more than 40 dark pools, with no single stock exchange having an overall market share greater than 20% of the consolidated volume for all National Market System (“NMS”) stocks. We also now have approximately 33.2% of trading volume occurring on dark pools, broker pools, and other “non-exchange” venues.

The rise in the number of trading venues — on exchanges and off-exchanges — and the fragmentation of trading volume, and the reliance on automated systems have increased the potential for system failures to spread quickly and affect the entire market.

Market Failures

Unfortunately, we have already seen too many instances of system-wide problems that have affected our capital markets. Some of the better known examples include:

• The flash crash of May 2010;
• The series of problems associated with the initial public offering of Facebook, Inc. in May 2012;
• Knight Capital Group Inc.’s $440 million trading loss in August 2012; and
• The issues associated with BATS Global Markets Inc.’s initial public offering in March 2012, as well as the recent admission by BATS that, for a period of more than four years, its computers for two equity exchanges and an options platform allowed trades to take place at prices that violated the Commission’s regulations, which seek to ensure that investors receive the best price.

Market failures of this magnitude and significance are unacceptable. Moreover, I am also concerned about reports that indicate that the recent market events caused by technology-related issues may have eroded investor confidence and impacted the stability and price-discovery function of our capital markets. According to recent data, since the May 2010 flash crash, investors have pulled over $377 billion from mutual funds that invested in U.S. stocks. While there may be competing reasons for these outflows, there are credible reports that indicate a causal connection between the flash crash, and the subsequent market disruptions, and the outflows of investor funds away from the equity markets. It is clear this is an area that requires regulation and oversight that result in transparency, stability, and accountability, and it is the Commission’s responsibility to propose and adopt solutions.

Although I recognize that there has been a steady uptick in the stock market over the past few months and that, according to recent media reports, some investors seem to be returning to the equity markets, I’m concerned about adverse effects that may result from future disruptions. To that end, I believe that we need to be proactive to prevent future market disruptions due to breakdowns in back-office systems and technology. It is not enough to merely respond to a past crisis, without solving the broader market stability issues.

Clearly, a reliable and robust market infrastructure is a critical component for protecting investors and ensuring fair, orderly, and efficient markets for all market participants. It is particularly disturbing that market professionals are losing confidence in our markets. A recent survey found that 26% of capital markets professionals had “very weak” confidence in our current market structure, up from 3% in May 2010.

Regulatory Measures to Promote Market Stability

Addressing the weakness and instability in the market structure must be a top Commission priority. To that end, the Commission must proactively develop an oversight and regulatory structure that is aligned with the purpose of serving the needs of investors.

In response to some of the market dysfunction, the Commission has adopted a number of regulatory measures designed to limit the broader impact of deficiencies and shortcomings of our market structure, such as single-stock circuit breakers, the ban on stub quotes, and rules banning naked access, and requiring pre-trade risk controls. In addition, the Commission has also adopted rules that require large traders, many of whom use high frequency trading strategies, to identify themselves so that the Commission can better monitor and analyze their trades.

While these measures are positive steps in the right direction, the Commission needs to do much more to update our regulatory framework and ensure that our capital markets develop and maintain systems with sufficient capacity, integrity, resiliency, availability, and security.

I recognize that our capital markets are very complex and inter-related, and all of the proposed solutions to the instability in our capital markets must keep this in mind. However, to not act is to turn a blind eye to the market instability that has had significant negative impact. The impact is compounded by the fact that our capital markets are complex and inter-related.

There is a great deal about our capital markets that we can all be proud of; despite its limitations, the U.S. capital markets are among the most developed, competitive, transparent, liquid, and vibrant capital markets in the world. Our capital markets truly set the standards for others to follow. The recent market malfunctions, however, have raised important questions that must be addressed to make sure that U. S. markets are working effectively — and, in particular, that they are working for retail investors. Some of these questions include:

• Have liquidity and price discovery been impacted by the flow of stock trading volume to off-exchange, and less transparent trading venues?
• Are exchanges, ATSs, and dark pools on a level playing field?
• How has the operating model of exchanges been influenced by their transformation from not-for-profit organizations to for-profit companies?

These questions crystallize issues that I cannot address in the limited time I have this morning, but I do want to highlight a few market stability measures that deserve serious consideration. These include:

• The Implementation of a Mechanism To Limit or Shut Down Orders — Such as a “Kill Switch”; and
• Live Simulations and Robust Testing of Business Continuity Plans for Trading Software.

The Implementation of a Mechanism To Limit or Shut Down Orders — Kill Switches

Just a few months ago, on October 2, 2012, the Commission conducted a roundtable entitled “Technology and Trading: Promoting Stability in Today’s Markets (“Roundtable”). The Roundtable examined the relationship between operational stability and integrity of the securities markets and the ways in which market participants design, implement, and manage complex and inter-connected trading technologies.

During the Roundtable, there was a robust discussion as to whether there should be a regulatory mandate to implement a mechanism referred to as a “kill switch.” The term “kill switch” is used to refer to mechanisms pursuant to which one or more limits could be established by a trading venue so that if a participant exceeded those limits, the trading venue could stop accepting incoming orders from the participant — in essence, a “kill switch” that would stop further trading.

I believe that the concept of a kill switch designed to catch extreme events and prevent market disruption is an idea that should be developed. Although I do not believe that any one individual safeguard will comprehensively eliminate risk in all situations; I do believe that a multi-layered approach to market structure with multiple, independent, coordinated, and overlapping risk checks are critically important, and that a kill switch would be one such important safeguard.

Live Simulations and Robust Testing of Business Continuity Plans for Trading Software

I also believe that the Commission should mandate robust and routine testing of trading software at exchanges and other liquidity centers. This need was amply demonstrated by Hurricane Sandy. When Hurricane Sandy hit New York City in late October 2012, all U.S. stock trading closed for a period of two days. Despite having existing business continuity plans that would have allowed for substantial trading to continue in the wake of a hurricane, trading was halted. Hurricane Sandy was predicted to hit the New York City area for a number of days prior to coming to shore, which would have been an adequate trigger for these plans. Yet, that is not what happened.

Specifically, two days before the storm, on October 27, 2012, the securities industry conducted its annual test for trading firms, market operators, and their utilities to operate through an emergency by using backup sites, backup communications, and disaster recovery facilities. The test did not uncover any issues that would preclude the markets from opening with backup systems, if they chose to. However, the markets did not open, reportedly due to concerns about their operational effectiveness and, more importantly, the physical safety of individuals during the storm. Still, it is important to note that the business continuity plans in place did contemplate the possibility of a hurricane, and many of these plans anticipated the need to operate from off-site facilities. It is clear that these business continuity plans were not considered sufficiently robust to keep the markets open. I believe that market participants must be confident enough in their back-up plans to utilize them when the needs arise. Perhaps one way to increase the confidence in the effectiveness of a business continuity plan is to require that such plans be actually implemented in a live or near-live trading environment on a periodic basis.

In my mind, there’s not much difference between failing to have a business continuity plan and having a plan that you’re not confident enough to use. Hurricane Sandy should serve as a warning sign. It is not enough to have the false comfort of a business continuity program on paper. It is critically important for entities to robustly test their contingency plans and be prepared to use them.

In addition, I believe that we need a regulatory scheme that would extend the SEC’s Automation Review Policy (“ARP”) to include additional market participants such as alternative trading systems, plan processors, and clearing agencies. As many of you know, the SEC published the ARP as a policy statement in 1989, largely in response to the 1987 market crash. Noting the impact that systems failures could have on public investors, broker-dealer risk exposure, and market efficiency, the Commission expressed its view that SROs, on a voluntary basis, “should establish comprehensive planning and assessment programs to test systems capacity and vulnerability.” Although this is currently voluntary, the Commission is currently working on a rule proposal that will make this process mandatory, and I hope that we soon act on the proposal. The many malfunctions have amply demonstrated that a voluntary environment is not acceptable.

I have been asked how codifying guidance from 1989 can be the answer to the market structure issues of 2013. The question is a good one. Codifying the ARP guidance is only a first step towards addressing the concerns related to market structure. It by no means is the whole solution. Moreover, in codifying the ARP guidance, it is critical that the Commission propose rules that create transparency, accountability, and integrity in order to make sure that market participants develop policies and procedures that are reasonably designed to ensure that their systems have the levels of capacity, resiliency, availability, and security that will allow them to operate in an appropriate manner. If the ARP guidance is not codified in the right way, neither regulators nor the investing public will have the right information to provide a clear assessment of whether key market stakeholders are addressing their system compliance issues.

Conclusion

As the country’s capital markets regulator responsible for protecting investors, maintaining fair, orderly, and efficient markets, and facilitating capital formation, the SEC must be at the forefront of proactively addressing changes in our capital markets structure, and not merely responding to events that have occurred. The challenges posed by large, automated, complex, and fragmented trading centers are real, and solutions cannot be put off to an unspecified future for consideration. Our investors deserve a proactive regulator.