Market Structure Insights


IN NOVEMBER, THE FINANCIAL Industry Regulatory Authority issued a Regulatory Notice to remind brokers of their existing best- execution obligations.The regulator also went on record with some new thoughts about how best execution should be pursued in light of new technologies and practices. Specifically, it clarified that member firms consuming proprietary market-data feeds to support their proprietary trading activities should also be using those feeds to route customer orders.

It hinted strongly that it prefers to see brokers conducting order-by-order reviews of execution quality for bigger orders, despite the fact that periodic, batch-style reviews are still permissible. It expects brokers to pay more attention to concepts like latency and information leakage when assessing execution quality, and it cautioned that trading filters must be used in a way that does not compromise the customer’s execution quality.

Viewed in isolation, FINRA’s guidance represents an overdue raising of the best-execution bar for market participants who have grown increasingly sophisticated over the past several years. Viewing it in isolation isn’t practical, however, because the Securities and Exchange Commission is also weighing in on the topic of best execution.

In July, the SEC proposed new order-handling disclosures for customer orders exceeding $200,000 in value. Eight months earlier, around the same time that FINRA issued its guidance, the Commission proposed that Alternative Trading Systems be required to divulge additional information about how they operate. Both proposals are designed, in part, to produce information that will allow brokers to pursue best execution more intelligently. We like the concept of pairing new disclosures with an obligation to review them. It’s likely to improve the institutional investor’s trading experience on the margin.

However, this approach to policing best execution – what we call the “disclose-and-review” approach – probably won’t dramatically improve execution quality for institutional investors the way it did for retail investors 15 years ago.

As we discuss in more detail below, human judgment still plays a big role in the execution of an institutional order. Using data to gauge the quality of those judgments isn’t as straightforward as using it to judge the efficacy of a retail order- routing process that’s completely automated.


In April 2001, the National Association of Securities Dealers informed its broker-dealer members in a lengthy notice that “regular and rigorous” reviews of execution quality for smaller orders were necessary to demonstrate compliance with its best- execution rule. Just six months earlier, the SEC adopted Rules 11Ac1-5 and 11Ac1-6, now known as Rules 605 and 606 of Regulation National Market System. Rule 605 requires market centers to disclose execution-quality statistics like fill rate, effective spread, and speed of execution, and Rule 606 mandates that brokers disclose the venues to which they route significant numbers of small-sized customer orders (“small” is defined as below $200,000 in value).

Academics seem to agree that Rule 605 improved market quality for retail orders. Once retail brokers could see which venues were giving them better execution quality (more-frequent price improvement, faster executions, etc.), they could reward those venues by sending them more orders. The venues that weren’t performing well could either improve their metrics or lose out on order flow. Either way, the retail broker, and by extension the retail investor, was getting better service.

As for Rule 606, it currently requires brokers to divulge only surface-level information about their routing practices. Frankly, 606 disclosures aren’t terribly useful in a vacuum. When combined with Rule 605, however, they likely encouraged brokers to actually divert order flow from “bad” venues to “good” venues. Brokers choosing not to do so would have had to justify why they were sending orders to venues offering subpar execution quality.

Meanwhile, the NASD stated explicitly in its guidance that brokers must conduct regular and rigorous reviews of execution quality for small-sized orders every quarter, if not every month, if not on a transaction-by-transaction basis. The reviews must cover venues to which the broker does not route orders in addition to those to which it does, making the 605 data indispensable. How else could a broker assess execution quality at venues to which it did not route orders?

The net effect of the 605 and 606 disclosures and the NASD- mandated reviews was to force brokers to code their order-routing systems to preference the best-performing venues. Needless to say, retail investors benefited from this.


Broadly speaking, we expect that the initial stages of disclose-and-review will play out similarly for institutional orders. The new disclosures required of Alternative Trading Systems will flood the marketplace with new information about how they operate, and the new disclosures required of brokers will expose order- handling procedures that seem improper or unsophisticated in light of the new information. FINRA’s recent guidance will give further impetus to brokers to make use of the information.

The end result of all the data-crunching and analysis should be a net improvement in execution quality for institutional orders. Smaller institutions that lack the resources to attract top-level service from their brokers should be particularly well served.

If there’s a danger embedded in the disclose-and-review approach, it’s the temptation to accept the data at face value without discounting it for its potential flaws, omissions or biases. While the routing logic for retail orders is fairly straightforward – there’s almost always enough liquidity within the network of electronic venues to satisfy the full size of a retail order in a single route – the routing logic for bigger institutional orders is more complicated.

According to a recent Greenwich Associates survey, 60% of all US equity dollar volume executed on behalf of institutional investors in the first quarter of 2016 was executed by a single-stock or portfolio sales trader. While many of those orders were likely executed algorithmically, some may have been executed manually and some may have been executed algorithmically and manually. The manual component – think orders executed through direct market access or crossed “upstairs” – involves human judgment that won’t be measured by the routing data that brokers will have to disclose to clients and the public. In fact, even the orders that are executed algorithmically are influenced by the sales trader’s decision of what algorithm to use and what parameters to select. These are clearly “best execution” decisions, but the quality of those decisions won’t show up in the data.

What’s more, there’s a real risk that differences in the data across brokers will be misinterpreted as resulting from differences in automated order-routing tactics, when in fact they might result from differences in judgment about how to best complete an order. Broker A might have a different style or philosophy than Broker B, or Broker A may just find itself working orders with different characteristics or in different market environments that call for different judgments upstream from any routing decisions.


The regulators’ disclose-and-review approach should provide valuable insight into how brokers handle and route institutional orders. And the pressure that brokers will be under to make their own data look as good as it possibly can should have a salutary effect on execution quality for institutional investors. That said, the disclosed data will only provide insight into, and will only pressure brokers to improve their order-handling procedures around, how cannily or efficiently the broker executes after a high-level judgment about how to complete the order has been made. As such, disclose-and-review will only give institutional investors and their brokers part of the best execution story.

DISCLOSURE: This publication is prepared by Weeden & Co.’s trading department, and not its research department. This publication is for information purposes only and is based on information and data from sources considered to be reliable, but it is not guaranteed as to accuracy and does not purport to be complete and are subject to change without notice. This publication is neither intended nor should be considered as an offer or the solicitation of an offer to sell or buy any security or other financial product. Nothing contained herein is intended to be, nor shall it be construed as, investment advice. Information contained herein provides insufficient information upon which to base an investment decision. Any comments or statements made herein do not necessarily reflect those of Weeden & Co. LP or its affiliates. © 2016 Weeden & Co. LP.


Editor’s Note

When was the last time the functioning of equity markets made news?

Dear Fellow STA Member

A note from outgoing Chairman Jon Schneider.

Shining Light on Dark Pools

Venue operators face scrutiny into how they conduct business.

November 8 Looms Large

Clinton or Trump? In a very unconventional election, the Republican candidate is the one feared by Wall Street.

EMSAC Comes of Age

Market structure advisory group ramps up activity, keeps SEC busy with recommendations.

CAT Moves Forward Slowly

The industry is on board with the concept but troubled by some details.

Equity Exchanges Evolve

Competition, fragmentation, and technology are among primary themes for listed-trading venues.

Market Structure Insights

Assessing the Disclose-and-Review approach to best execution. By Andrew Upward, Weeden

Spotlight on HFT Dims

High-frequency trading may not have quite gained acceptance — but at least the pitchforks have been put aside.

T+2 Draws Closer to Reality

But will new technologies complicate the migration?

Maker-Taker Under the Microscope

Has the nearly 20-year-old exchange pricing model outlived its usefulness?

Re-Imagining Reg NMS

Landmark equity-market ruleset is showing its age; how to update?

Options Auctions Vex Market Makers

Sidelines transactions diminish already-challenged displayed liquidity.

The Market Data Battleground

Data fees are ridiculously expensive… at least that’s what the brokers say.