By Patrick Flannery, Co-Founder and CEO, MayStreet
For as long as there have been markets, there’s been a demand for information about them. In the US, the rules that govern equity market data were built in the 1970s, and it’s time for an upgrade to them. Not only is it long overdue, but doing so will empower modern technology to provide market participants the information they need while minimizing risks and abuses.
But before we focus on where we need to go, let’s take a look at where we are and how we got here.
In the early 1970s, the Securities and Exchange Commission decided that the patchwork of different exchanges providing different quote and trade information meant that market participants didn’t know where to get the best prices. It then began to adopt rules that would later become the CTA/CQ and UTP Plans, which ultimately paved the way for the introduction of the SIPs. The purpose of these “public” market data streams – both then and now – is to provide market participants with a real-time view of essential information about quotes and trades at a reasonable cost.
Through them, market participants were able to easily view the best bid and ask from each and every US equity exchange in a single, easily consumable data feed, creating a level of transparency and efficiency that was unheard of at the time (and remains the envy of other advanced markets around the world). But today – nearly 50 years after their introduction – the SIPs face major questions about their relevancy.
Simply put, the public market data streams no longer serve as intended. They are too slow (at least in relation to the proprietary data feeds provided by the exchanges). They don’t provide sufficient information to meet the needs of many investors. They are expensive, and their fee structures are extraordinarily complex. None of that’s an accident, as the exchanges generally profit handsomely from the current system.
As mandated under the first CTA Plan (and reaffirmed with the passage of Regulation NMS in the mid-2000s), the SIPs are run by the SROs, which include FINRA and the growing number of exchange operators (NYSE, Nasdaq, Cboe, IEX and now MEMX, LTSE and soon MIAX). Each of the exchanges offers their own data feeds, which compete with the SIPs. Some of these “proprietary data” feeds offer more information, faster, and for a hefty price. Further, in addition to overseeing the SIPs, NYSE and Nasdaq have business units that also serve as administrators and technology processors for the CTA/CQ and UTP SIPs, respectively.
Voting powers over the SIPs are allocated based on the number of exchange medallions held – given the nine combined markets operated by NYSE and Nasdaq, those companies have 50% of the total votes on the Operating Committees that supervise the performance of their administrator and processor businesses. A tangled web for sure.
The SEC has spent the last three years focused on how to best modernize the SIPs from a variety of perspectives. The SEC’s October 2018 market data and market access roundtable organized a wide swath of market participants who nearly universally agreed that reforms to the SIPs are long overdue. Perhaps as a result, Trading and Markets Division Director Brett Redfearn – who, given his previous job as J.P. Morgan’s Global Head of Market Structure, arguably understands the intricate details of the SIPs as well as anyone – has made it a priority to address many of the SIPs’ problems.
Earlier this year, the SEC has issued an order directing the exchanges and FINRA to modify the governance structure for overseeing the SIPs. In particular, the SEC is pushing on the SROs to give investors and brokers a say in how the public market data stream that they use and pay for is run. The order went into effect in May and the SROs continue to work on the mandated changes. When finally implemented, which could take some time, it will make the SIP governance more balanced. But it is not a panacea. The SIPs themselves need to be better.
As Mehmet Kinak, T. Rowe Price’s Global Head of Systematic Trading and Market Structure, bluntly put it when speaking on one of the market data roundtable panels in 2018: “If a broker is routing using SIP data, they are not routing my flow. They can route someone else’s, but they’re not eligible to get my flow, period.” In our experience working with the largest banks and buy sides on the Street, Mr. Kinak’s view is not unique.
To help address the concerns expressed by Mr. Kinak and many others in the industry, the SEC in February proposed to fundamentally reimagine what the SIPs should be. The SEC’s SIP Proposal would add content to the SIPs, including several layers of depth-of-book, auction information and odd-lots (which are a large and growing segment of orders and trades).
It also, importantly, opens up the SIPs for competition. Despite the fact that the Plans are supposed to be competitively bid, they generally haven’t been. The SIP Proposal would, if adopted, allow competing providers to construct “SIPs” with different characteristics, which could perform the governmental functions of the current SIPs.
We at MayStreet whole-heartedly support this reimagining of the SIPs, and believe that—given today’s technology—the market can create a solution that’s better performing, lower cost to operate and consume and potentially paves the way for new delivery models (for instance, “near-time” data sent via the cloud as one lower-cost option). “Let the invisible hand of competition and market efficiency do the work of bringing down the costs of market data and, with proper regulation, competition will do it far better and faster than any centrally planned system,” was how Roman Ginis, a former buy-side trader and founder of IntelligentCross, characterized it in his comment letter on the proposal, and that’s a sentiment with which we certainly agree.
To be sure, there are legitimate questions a distributed SIP model raises that the industry must think through. For example, how should we handle the added complexity of potentially having multiple NBBOs? Or how do we ensure that data providers (whether exchange-related or not) all receive data at the same time, place and format to ensure a consistent starting line for all? Valid concerns, but in our opinion none are insurmountable and all can be solved with smart regulation by the SEC. The Healthy Markets Association, among others, provided suggestions for dealing with these issues that we believe are worthy of consideration.
So as the SEC sorts through the dozens of comment letters received on the February proposal, allow us at MayStreet to make a proposal of our own: let all of us in the market data trenches—the vendors, brokers and investors who pay for and work with the SIPs and direct feeds each day, as well as the exchanges themselves if willing—come together over the next few months and present to the SEC a serious proposal for how to move forward. We’re happy to pick up the mantle on this effort or work through existing industry groups like FIF, SIFMA, STA or Healthy Markets—as no doubt, many folks already are. But no matter what, we believe that the voice of market data consumers has been shut out for too long and that must be rectified.
With the benefits offered by today’s technology, the power of competitive forces and industry-wide participation, we believe significant change that can reduce the cost and improve the quality of market data is well within reach. While SIPs have generally served the industry well these past four-plus decades, everything has a shelf life. We’re excited about the potential to rethink this vital component of our market structure, and look forward to working with our colleagues across the industry to do so.