Fears grow as exchanges up sales of data

Money managers and institutional asset owners are concerned that the volumes of equity market data being collected by exchanges could be used to find information on institutional trade strategies and volume, leading to front-running or other problematic behaviors.

The concern was highlighted in December when Nasdaq Inc. withdrew an application to the Securities and Exchange Commission to introduce Intellicator, a market data product that would "distill options data into calculations and indicators" and "reverse-engineer factor calculations to obtain transaction-specific information not otherwise available on the exchange's data feeds," according to Nasdaq's filing.

The Securities and Financial Markets Association, which represents ​ brokers, banks and money management firms, and buy-side participants, said in comments on the Nasdaq proposal to the SEC that such reverse engineering would provide those who purchased Intellicator with trading information, including details on block trading behavior, that could be used in market manipulation.

Nasdaq in a Dec. 22 letter to the SEC denied those claims but said it would refile an application to the SEC on Intellicator once it can "further educate market participants and potential customers about the functions and capabilities … and to consult with interested firms about any concerns they may have, as well as possible modifications to the product."

Representatives at Nasdaq, New York Stock Exchange and Cboe Holdings Inc.'s Bats equity trading unit, all of whom sell market data, would not comment for this story. The Intellicator controversy has highlighted the concern among institutional investors that, as exchanges increase the kinds of market data sold to traders to increase overall revenue, those products could display trading information that large investors are trying to hide or already have obtained internally.

"I do understand where the complaints are coming from," said Steven Glass, president and CEO of Zeno Consulting Group LLC, a Bethesda, Md., consultant to pension funds on trading issues. "Managers who are running institutional money break up their trades to disguise what they're doing. Some of these data tools could enable people to identify when block trades were occurring. (The exchanges) would argue they were being transparent, but on the other hand, the fact that this data could help people jump over an institutional order, that certainly would be a concern for our clients. It's worth shining a light on."

Valerie Bogard, New York-based equity analyst at TABB Group LLC, said institutional investors are "certainly" going to be cautious about anything that could lead to information leakage, particularly as exchanges continue to create more revenue streams from data sales as execution revenue declines. "Trading volumes are down, so exchanges are looking at creative ways to sell data," Ms. Bogard said.

Issues of market data sales revolve around two main themes, said John Ramsay, chief market policy officer at IEX Group Inc., a New York-based stock exchange: data costs at a time when traders need more of it to maintain best execution, and what the data is used for. While the rising cost of market data has been an issue for years, Mr. Ramsay said, what information is being sold is a big issue as well, as evidenced by the clamor against the proposed Nasdaq product. IEX does not sell its market data, Mr. Ramsay added.

"The reason there's so much controversy about Intellicator is that Nasdaq said the feed could be used to reverse engineer how a specific type of participant was trading," said Mr. Ramsay. "Large buy-side institutions want to trade in size without tipping off the market, but this tool would enable traders to know when those large buy-side firms are in the market. It's designed for people looking to pick off investments. The (Nasdaq filing with the SEC to withdraw the data products) indicated that their customers raised concerns about this."

(Nasdaq's response to the SIFMA letter is available on the SEC's website.)

What needs to be addressed is what "clear line" would mark the point at which market data could benefit some and hurt others, said David Weisberger, head of cryptocurrency and equity markets at ViableMkts, a trading technology and market structure advisory firm in New York.

"The (Food and Drug Administration) could sell information on their rulings in advance," Mr. Weisberger said. "They could make a lot of money on that, and fund a lot of programs, but is that fair? The government could also profit by selling economic data early to trading firms willing to pay for it, but that would not be fair either. In those cases, there is a clear line between what can be done and what can't. There should be a similar line with exchanges, and no one really knows where regulators stand on this … The bottom line is that there should be a comprehensive policy on data for all asset classes that describes what should be publicly available and what can be sold."

Sources interviewed for this story agreed the SEC has not been thorough in its assessment of exchanges' market data proposals. The SEC generally "rubber stamps" exchange data product proposals, said Tyler Gellasch, executive director of Healthy Markets Association, Washington, a non-profit group of money managers, brokers and pension funds with a combined $1.5 trillion in assets held or under management that lobbies on trading and market structure issues. "Information leakage and cost issues all stem from a fundamental problem — for-profit firms are writing these rules and the SEC is approving them," Mr. Gellasch said.

Among Healthy Markets Association's 10 members are the $355.5 billion California Public Employees' Retirement System, Sacramento; PSP Investments, which manages the assets of the C$135.6 billion ($109.2 billion) Public Service Pension Investment Board, Montreal; Janus Henderson Group PLC and Brandes Investment Partners. In a Jan. 17 letter to SEC Chairman Jay Clayton seeking to revamp the SEC's procedures for approving exchange market data proposals, the association said for-profit exchanges "have been able to exploit their essential role in the market infrastructure to add complexity and costs to a broad swath of market participants. And we found that almost no information related to the volumes and impacts of these fees is public."

John Storrs, Chicago-based head of equity trading at Northern Trust Capital Markets, a broker for other money managers, said that as technological advances have led exchanges to produce more trading data, so too has the need of brokers, money managers and end clients to monitor what data tools are being introduced. Northern Trust uses a wide array of analytics to constantly check for anomalies in the equity markets, but also, "We talk to our clients every day about what are the best tools out there to determine whether there's not too much information out there," Mr. Storrs said. "For people on the front lines (of trading), it's up to them to see what (data providers) are doing and say, 'Hey, there's something wrong here.' That's why we keep open communication with clients on a daily basis."

While institutional money managers generally will seek more data to help achieve best execution, Jeffrey A. Levi, Darien, Conn.-based principal at money manager consulting firm Casey Quirk, a practice of Deloitte Consulting LLP, said they should specify what kind of data they need based on the capabilities they want, their trade policies and how much alpha generation they seek.

"Trading can be an important source of alpha, that's one dimension" Mr. Levi said. "But data could also be used for mining, to gain an edge by identifying patterns of all traders' activity. It's there where it could be concern. There is no silver bullet yet. (Curtailing market manipulation through exchange data mining) will come down to a lot of discussions with regulators on what to do. There will be difficulty in creating boundaries. Sources of data are growing from all kinds of sources, with consequences across the board."

Mr. Levi said the unanswered question among money managers is which do they prefer, alpha or trading anonymity? "That's a good question. I don't know the answer to that. I'm not sure you can perfectly police things so that both remain, but you can draw some limits."