Introductory Remarks at SEC's Market Technology Roundtable
Chairman Mary L. Schapiro
U.S. Securities and Exchange Commission
Oct. 2, 2012
Good morning. Thank you to all of the panelists for taking time to share your thoughts with us on market technology. And thank you to those who have already written in with your comments. You have given us a number of very thoughtful recommendations.
To an extraordinary extent, the stability of our securities markets is tied to the technological infrastructure of those markets.
As with virtually every industry, technology brings many benefits. And our markets are no different. Thanks to technology, our securities markets are more efficient and accessible than ever before.
But we also know that technology has pitfalls. And when it doesn’t work quite right, the consequences can be severe.
Just imagine what can happen:
- If an automated traffic light flashes green, rather than red.
- If a wing flap on a plane goes up rather than down.
- If railroad track switches and sends the train right, rather than left.
Similarly, there can be significant consequences for technological errors in our markets as well — trading can be disrupted, investors can suffer financial loss, firms can be imperiled, and confidence in our markets broadly can erode.
Today’s roundtable will help us think through the issues and the steps we need to take to ensure that our markets remain the most robust, efficient, and stable in the world.
There are two basic concerns we need to focus on that are highly interrelated — these are:
- First, the structure of our markets, such as multiple execution venues, the presence of high frequency trading, dark pools, and the like.
- Second, the infrastructure of our markets, as in the technology that undergirds trading activity.
To provide some perspective — in January 2010, I asked the staff to begin a comprehensive review of the equity market structure. It was a review that included getting views on everything from the impact of high frequency trading, to the continued rise of dark pools, to the complexities of a multi-venue market system.
The focus was not so much on the infrastructure of our markets but on the way the markets and market participants operate and behave.
Four months later, when disorderly trading activities in the S&P e-mini market spread to the equities market, causing what is now known as the Flash Crash, we — as an agency — were well-positioned to respond. Working with the exchanges, we quickly put in place a series of measures that have since helped to reduce the likelihood of another event like that from occurring.
Within days of that event, I summoned the heads of every exchange to the SEC to hammer out common-sense approaches to bolster our markets.
And as a result of our efforts:
- We now have in place single-stock circuit breakers to prevent stocks from falling too far, too fast, and we have approved a more advanced limit up/limit down mechanism to limit excessive volatility.
- We now have in place a ban on stub-quotes and rules clearly defining when a trade can be broken so as to help avoid circumstances that can lead to disorderly trading.
- We now have in place rules banning naked access and requiring rigorous pre-trade risk controls designed to help mitigate disruptive trading at the source.
- And we now have rules requiring large traders, many of whom use high frequency trading strategies, to identify themselves so that the Commission can better monitor and analyze their trades — a process that other regulators overseas are beginning to emulate.
Additionally, and perhaps most importantly, we have adopted a rule that requires SROs to develop plans for the first ever consolidated audit trail — a feature that will allow regulators to surveil and reconstruct trading across platforms.
But there are issues around market structure and the conduct of market participants that we should further examine, including the high volume of cancellations, a proliferation of order types, transparency, high frequency trading generally, potentially manipulative trading strategies, and data latencies for public investors — to name a few. These issues still require attention and we are committed to addressing them.
But today’s roundtable will focus more specifically on infrastructure not only because of its importance, but also because I worry that this issue is at risk of being lost and subsumed by the broader debates regarding market structure. After all, issues that get lost often do not get resolved. And these matters of infrastructure are essential to any holistic approach to improving how our markets operate.
Consider for a moment the IPO of BATS on its own exchange, and the IPO of Facebook on the NASDAQ exchange. Though there are many views regarding the fragmented nature of simultaneous trading across multiple venues, I believe these IPO events evidence a very different set of concerns. Both events involved one of the few single-exchange processes that remain in an otherwise fragmented market — namely, building a single order book and crossing trades at a single price to open trading for a new public company.
In the case of BATS, it was a flaw in new software code designed to conduct a corporate IPO auction. That mistake caused the matching engine for tickers in a certain range to enter into an infinite loop, making these tickers, which included the symbol for BATS itself, inaccessible on BATS.
In the case of NASDAQ, the IPO software was designed to accept cancellations submitted while the final IPO price — or the Cross — is being calculated. Cancellations received during this time changed the order book. By design, the system recalculated the final IPO price to factor in the new state of the book. But again, changes were received before the system could print the opening trade, which resulted in additional re-calculations. This condition persisted, resulting in further delay of the opening print.
These single-exchange problems are not a result of complexities or fragmented markets, but rather a result of more basic technology 101 issues.
Consider as well the events this summer with Knight Capital — a trading firm that had just installed trading software that was intended to send orders to the NYSE’s new Retail Liquidity Program. Instead, the software wound up sending “a ton of orders” into the market. As the market data that morning revealed, the software did not create patterns of rapid orders and cancels. Rather, the data showed a massive amount of orders resulting in executed trades that caused Knight Capital to accumulate significant and unwanted positions.
This type of problem, as with the IPO mishaps, was the result of basic technology 101 issues.
Events like these demonstrate that core infrastructure and technology issues can be problematic in any market structure.
However, though for today we are focusing on infrastructure issues, it is important to recognize how the overall structure of our markets can affect how our infrastructure is designed and implemented. For example, we have a very competitive market environment in which rapid innovation and speed-to-market may compete with diligent testing and validation of the technologies that support such innovation.
Our multi-venue, interlinked market structure also means that an infrastructure failure by one party or at one venue may cascade into other venues and affect many other parties. And of course the inherent speed of trading, which itself is partly a result of the competitive nature of our markets, means that even small, short-lived infrastructure issues can cause drastic harm.
To be sure, several of the measures we have already approved have helped to strengthen our markets even in the face of potential, and inevitable, technological errors.
Indeed, several of the post-Flash Crash reforms such as the revisions to the clearly erroneous rules helped limit the impact of the Knight Capital episode on other market participants.
But limiting any harm resulting from technological errors is not as good as preventing the error in the first place — which is why we have instituted clear rules that require firms with access to our markets to have controls in place to reduce the chance of such errors.
But perhaps the strongest message from the Knight Capital episode is that the party committing an error may very well l end up bearing a massive financial loss. That, more than anything, sends a wake-up call to the entire industry.
Nonetheless, our concern is not whether a single firm might fail, but whether it causes collateral damage to investors and their confidence in the integrity and stability of our markets.
So I am pleased that the industry has been working overtime in the aftermath of the Knight Capital episode to address these issues — and I am pleased our roundtable has spurred such discussion.
By focusing on the underlying nature of these incidents, and hearing, as we are today, from experts in technology, I hope we can address these issues in an efficient, effective, and expeditious manner.
I now will turn it over to Robert Cook, Director of the Division of Trading and Markets, and his staff who will serve as moderators of today’s discussion.