The Benefit of Hindsight and the Promise of Foresight: A Proposal for A Comprehensive Review of Equity Market Structure

Commissioner Michael S. Piwowar

U.S. Securities and Exchange Commission

London, England

Dec. 9, 2013

Thank you Dan [Waters] for that kind introduction and the invitation to join you today.  It is nice to be among such a large and distinguished group of market structure experts from both the private and public sectors.  As a former academic researcher in the area of market microstructure, who has also spent time in the private sector and in government, this conference is a great setting for me to give my first public speech on equity market structure as an SEC Commissioner.  Over the next several months, I plan to speak about each of the three parts of the U.S. Securities and Exchange Commission´s (SEC´s) core mission – protecting investors; maintaining fair, orderly, and efficient markets; and promoting capital formation.  Today´s speech covers the second part of our mission – maintaining fair, orderly, and efficient markets – and in it I would like to share with you my views on the need for a comprehensive review of equity market structure.  It is also fitting for me to be delivering a speech in a venue called "The Brewery" since my last name – "Piwowar" – literally means "Brewer" in Polish.

At the outset, I need to provide the standard disclaimer that the views I express today are my own and do not necessarily reflect those of the Commission or my fellow Commissioners. 

Let me begin with the Wall Street Journal´s colorful reporting on a sharp and sudden decline in the securities markets that happened one day in the month of May:

The stock market careened downward yesterday, leaving traders shaken and exhausted. … The drop took place on volume so heavy that the stock price ticker wasn´t able to finish reporting … transactions until 5:59 p.m., two hours and 29 minutes after the market closed.  This lag was the longest since … 1930.[1]

Other characterizations of the market conditions on that day are just as dramatic.  The market behaved "erratic[ally]" and "frenetic[ally]."[2]  Some stock prices dropped "relentlessly," with one "crash[ing] like a boulder pushed from a cliff."[3]  The volatility "terrified investors," "caused bewilderment at home and abroad," and had a "strong and immediate psychological impact."[4]  It is said that in the midst of the activity, market participants were having trouble keeping track of their own order executions.[5]  There was widespread speculation about what was going wrong. 

What may surprise you is that none of what I just described refers to the May 6, 2010 "flash crash" or any other recent market event.  Instead, this narrative is from 1962.  That is right – 1962.  On May 28, 1962, the U.S. markets suffered a so-called "break."  As it turns out, sudden and unexpected market disruptions, in other words "breaks" or "flash crashes," are not recent phenomena.  Flash crashes occurred well before the emergence of computerized algorithmic trading, high-frequency (or low-latency) trading, alternative trading systems, smart order routing systems, decimalization, and even the founding of the National Association of Securities Dealers Automated Quotation System (NASDAQ).    

Hindsight

Given the similarities between the recent "flash crash" and the 1962 "break," there is a lot that we can – and should – learn from the events of 1962 and apply to the markets today.  For instance, to the extent people are inclined to use the flash crash as a convenient pretext to malign high-frequency and automated trading, that overlooks what history shows us is clearly true – the advent of technology-driven trading has not singlehandedly corrupted our markets.  Speed may exacerbate problems, but there is no definitive evidence that it is the problem.  Moreover, we should not reject the possibility that speed may actually help mitigate problems once they begin.  In addition, it is hard to make the claim that fragmentation is the sole, or even primary, cause of market disruptions, since trading was much more concentrated in 1962 than it is today.

A significant part of what we can learn from the events of 1962 stem from an analysis that was set in motion before the market turmoil of that year.  In 1961, Congress mandated that the Commission undertake a study and investigation of the adequacy of the rules of national securities exchanges and national securities associations.[6]  That comprehensive review of securities markets involved a 19-month effort and resulted in a five-part, 13-chapter "Report of Special Study of Securities Markets of the Securities and Exchange Commission (Study)," including one chapter dedicated to the events of May 1962.[7]  The section analyzing the market break, while extensive, did not yield a conclusion about a precipitating cause (or causes).  Rather, the report stated that "[t]he history of the May 28 market break reveals that a complex interaction of causes and effects – including rational and emotional motivations as well as a variety of mechanisms and pressures – may suddenly create a downward spiral of great velocity and force.  This, in turn, may change the impact of various normal market mechanisms, and thus temporarily impair the market´s fair and orderly character."[8]  This sentiment rings just as true for our markets today as it did over 50 years ago.

As insightful as the findings of the Study are for us in understanding today´s markets, just as significant was the manner in which the review was conducted.  In fact, I believe that the report provides a useful framework for analyzing our current market structure issues.  The Study was intentionally broad, and constituted the most comprehensive review of markets in 25 years.[9]  Milton Cohen, who has been described by the Commission as a "seminal figure[] in the history of the SEC"[10] and the New York Times as a "lawyer who inspired market reform,"[11] was the director of the Study.  Separately from the Study, he published his own reflections on how it was structured and undertaken.[12]  Mr. Cohen lamented the "pitfalls and booby traps" of conducting a comprehensive market study, including the difficulty in drawing boundary lines around the topics to include.  He concluded, however, that when "studying rules and practices of the securities markets and the adequacy of regulatory and self-regulatory measures, one [can] hardly deal with a single facet … without also dealing with all or most of its other facets."[13]

It is noteworthy that the Study represented the work of a group comprised not only of SEC staff, but also included, among others, individuals from private law practice, academia, and industry.   In his post-Study musings, Mr. Cohen surmised that "[i]f the Commission members themselves were to have gone as deeply into each of these processes as the Special Study has done, either the scope of the Study would have had to be narrower, or the time longer, or the recommendations very much less specific."[14]

In contrast to the expansive Study, in the aftermath of the May 2010 flash crash, the staff of the SEC, along with the staff of the CFTC, produced a report that focused solely on what happened during that market event.[15]  The SEC staff did quality work on the narrow report to which they were assigned.  However, I find it troubling that the SEC has still not conducted a comprehensive review of market structure in light of the 2010 flash crash.  To the contrary, the SEC has effectively abandoned its Concept Release on Equity Market Structure that was published for public comment in January 2010.[16]  These are missed opportunities, and the Commission´s inaction needs to be remedied.

Foresight

Action is necessary because we face a market that is swiftly evolving, and we thus confront the challenge of making sure that our regulatory structure does not conflict with how our securities markets operate and how market participants interact with one another.  If we do not have a fundamental appreciation for the context in which our rules and regulations are applied, the Commission cannot be effective in carrying out all three parts of our core mission:  protecting investors, maintaining fair, orderly, and efficient markets, and promoting capital formation.  I therefore echo my colleagues at the Commission who recognize the need to review our current market structure[17] and reiterate my support for a comprehensive review of U.S. equity market structure.  The pressing question, then, is what I mean when I refer to a "comprehensive" equity market structure review.

When constructing an equity market structure review today, we have the benefit of Mr. Cohen´s wisdom.  He advocated for considering a broad range of topics, to gain a better perspective on how components of our markets interconnect.  I agree.  An examination of issues separately and/or successively will suffer the same drawbacks as the SEC´s "Market 2000" report.[18]  That report, while an exemplary product by the SEC staff, generated unwarranted expectations and made only "incremental" recommendations.[19]

A review cannot be focused narrowly on what may have caused the most recent market disruption or trading "glitch."  Instead, it is imperative that a market structure review cover a much larger scope.  Topics we must consider include, but are not limited to, market infrastructure (i.e., technology and interconnectivity of market centers), the classification and treatment of different types of market participants, undisplayed liquidity, exchange pricing models, off-exchange trading, self-regulatory organization oversight, and a Regulation NMS "regulatory lookback."      

Let me analogize to an effort many of you may be familiar with, and that some of you may have been a part of – the UK Foresight Programme (Foresight), which is designed to help the UK government think systematically about the future.[20]  Foresight brings together people from industry, government, and academia to perform in-depth, multi-year studies, as well as to develop shorter reports, that combine the latest in science and empirical evidence to inform policymaking.  Teams of subject-matter experts are assembled to focus on particular projects so that multiple topics can be researched simultaneously. 

One such Foresight project, completed last year, is the two-year study on "The Future of Computer Trading in Financial Markets – An International Perspective."[21]  The work involved over 150 leading academic and industry experts from more than 20 countries.  It yielded over 50 commissioned scientific papers and a final report that draws upon available science and empirical evidence to provide valuable advice to regulators about the potential risks and opportunities from the evolution of computer trading over the next decade.

The UK Foresight model is one that we should entertain in the U.S.  The SEC can benefit tremendously from collaboration with market structure experts from both the private sector and the academic world.  We can also cover a much wider range of topics – in other words be more comprehensive – and do so in more depth if we can leverage the resources of outside parties.

Setting Our Sights

To be clear, a comprehensive equity market structure review does not necessarily mean developing a grand plan for solving all problems, real or imagined, big or small.  In fact, the surest way to guarantee that nothing gets done these days in Washington, DC is to call for a grand plan or grand solution.  It does, however, mean we should put everything on the table for discussion.  And it is imperative that we undertake any review with a completely open mind rather than being predisposed to the status quo or prior suggestions for market reforms.

In addition, we must not immediately leap toward trying to determine the "optimal way" in which markets should be structured, lest any undertaking be doomed from the start.  In my view, the first step in a review should be to explore how it is that our equities markets have evolved to where we are now.  What has been the role of competitive forces in our marketplace, and how, if at all, have our regulations contributed to how our markets operate?  For example, we must question whether the market fragmentation and intermediation that many now decry as disadvantaging retail customers may be directly attributable to Regulation NMS.[22]  The Commission must be open to the possibility that any problems we identify are direct or indirect consequences of our own actions. 

Incorporating into a comprehensive review a retrospective analysis of regulations such as Regulation NMS is not an effort to roll back regulation or call into question the work previously done by the Commission.  In fact, it would be fully consistent with the expectations of an Executive Order recently issued by the President of the United States.  Independent regulatory agencies, such as the SEC, are expected to "promote retrospective analysis of rules that may be outmoded, ineffective, insufficient, or excessively burdensome, and to modify, streamline, expand, or repeal them in accordance with what has been learned."[23]  In other words, in order to move forward we must look back.

Furthermore, in a review of equity market structure we should not spend our time at the outset thinking about market structure concepts in the abstract – who is or who is not a so-called high-frequency trader, how fast is too fast, or how many trading venues are too many.  Rather, we should ask what incentives underlie the current market structure.  What drives the supposed "need-for-speed?"  Why are traders directing flow to so-called "dark pools" rather than "lit" markets?  These are not easy questions, but they are threshold ones that we must tackle.     

Once we identify the incentives at play – and I say incentives, plural, because I have no doubt there are numerous market-based and government-based incentives – we can then make choices about which alternatives may best facilitate competition on choice, pricing, and innovation.  We may disagree in those judgments, but that is the nature of policymaking and it should not prevent us from undertaking the comprehensive analysis that will serve as the basis for such policy judgments. 

An equity market structure review also must not be conducted in a vacuum.  As with anything we do, it is important that we hear the views and perspectives of the public to inform any market structure debate.  This is where the SEC will benefit to the utmost degree from data and research.  The agency has a talented staff of economists and other market structure experts who, particularly with the benefit of technology, can perform sophisticated analyses of how our markets work.  But that does not mean that we cannot benefit from the thoughtful work of academics and other market participants.  To the extent you have not done so already, I encourage you to visit the SEC´s Market Structure Data and Analysis website to see the extensive amounts of data on equity market structure that are available for your use.  The website is something the SEC rolled out this fall, and we are particularly proud of it.  Public users are able to download dozens of datasets to perform their own analyses and provide direct feedback to SEC staff.  I hope that this innovative project will allow us to gain a much richer understanding of market quality and liquidity dynamics, as well as the manner in which orders are sent to, and interact with, the market.

The comprehensive equity market structure review I am advocating is an effort that will take a couple of years, not a couple of months, to do right.  I am not suggesting, however, that we sit idly by waiting for the completion of this review before beginning to address important, discrete issues in our markets.  There are modest, temporary changes we should make in the interim that have the potential to improve market efficiency and in any event will provide us with information that can inform our overall review.  Let me give you one example. 

I believe we should move quickly to introduce a pilot program on increasing the tick size for small capitalization companies.  Frankly, I am disappointed we have not already done so.  By way of background, in July 2012, staff of the Commission submitted to Congress a report on decimalization, which was mandated under the JOBS Act.[24]  That report served as a predicate to a roundtable sponsored by the Commission earlier this year, which included a panel discussing a program for alternative minimum tick sizes.[25]  What we have learned through the study, the roundtable, and comments that have been submitted post-roundtable is that, while there are differing views on what benefits an alternative minimum price increment might have, there is great value in undertaking a pilot program.  Of course, like all policy actions, there is the possibility that a tick size adjustment could have adverse impacts.  However, the pilot program could include ongoing monitoring efforts designed to limit any such impacts.  Moreover, we cannot allow potential complications to paralyze us and prevent us from taking action, especially where it is likely that we will be statutorily mandated to undertake a pilot program if the Commission does not do so of its own initiative.[26]  Even if changing the tick increment does not have the intended effects, or any discernable effects at all, the results from the pilot program will serve as important data points going forward.     

Conclusion

To sum up, we should commence – in short order – a comprehensive, multi-year equity market structure review program, undertaken with the benefit of hindsight and the promise of foresight.  In doing so, we can set our sights on ensuring that the U.S. securities markets remain a world leader in market quality, efficiency, and fairness, which in turn will fuel global capital formation and economic growth.    

Thank you all for your attention.  I hope that you find today´s conference interesting and useful.