The revised Rule 606 has received significant attention recently, given its focus on the reporting of “hard data” on routing, including fees and rebates as well as other statistics. The revision is a significant step toward increasing transparency in an area of trading that can be fairly opaque. Brokers will now be required to disclose more about the fees and rebates they earn based on those decision, information that is often not readily available, and is available only at the brokers discretion. In addition, for orders that are providing liquidity, Rule 606 now requires that brokers report the average time between order entry and cancellation.[1]
As most traders are aware, fees and rebates provide brokers with incentives that may conflict with their clients. Because brokers typically earn those rebates and pay those fees, they have incentives to route in a manner to maximize rebates at the expense of execution quality. That’s not to say that brokers don’t have other incentives to trade in-line with their client’s interest. As discussed in a previous blog post[2], rebates can actually align incentives in certain circumstances. For example, a broker who routes to high rebate venues with sufficiently low fill rates run the risk of either a) not executing and missing out on a commission, or b) incurring “take” fees if they are unable to fill a “must complete” order passively and must resort to a marketable order. But with that said, there is strong evidence that broker routing isn’t always optimal. For example, an excellent academic paper by Robert Battalio, Shane Corwin, and Robert Jennings documents examples of such suboptimal routing.[3]
But at the same time, Rule 606 and aggregate reporting requirements in general can only go so far. For example, looking at the mix of make, take, and midpoint orders in the aggregate likely will not provide much useful information to most investment managers since the aggregate statistics mix together orders that may follow very different strategies. For example, an asset manager who uses a mix of VWAP, dark aggregation, aggressive “Sniper” algorithms, etc. may not find the information as clean as those who use only VWAP strategies. In the former case, the mix of take/make/midpoint will more likely reflect the mix of execution strategies than the underlying routing strategies of the brokers. The aggregation of diverse strategies also makes the “average time to completion” metrics of limited value since it depends on the trading strategy, the liquidity of stocks traded, etc.
Therefore, I think the best way to interpret the Rule 606 reports is as a canary in a coal mine – if there is something toxic, these reports will likely help identify those. It may also be useful in those rare cases where a client routes a nearly identical set of flow to multiple brokers and then can use the reports to identify any interesting cross-broker differences. For example, a client who uses only VWAP orders and routes similar mixes of orders to different brokers can make apples-to-apples comparisons across brokers. It can help also help a buyside trader put their performance into better context. For example, suppose the buyside desk sees poor performance at Broker A. If Broker A has a relatively low proportion of passive fills as documented in a 606 report and the rebates earned are large when those fills occur, the buyside desk may decide to focus their attention on the routing behavior as a possible cause of the poor performance.
Takeaways
I think it is pretty safe to say that the change to Rule 606 will be a useful addition to the information set available to buyside traders. More importantly, though, the fact that brokers are required to gather and report these data may make them more willing to provide buyside trading desks with these data in other reports and/or perhaps even provisionally on FIX messages.
With that said, one could argue that clients can already do this to a large extent already. If a buyside desk receives the make-take information in a FIX tag (or they can estimate it the fill is a make or take execution), the desk can use this information along with publicly available fee schedules from the exchanges to approximate the fees earned by the broker. However, such a method is often incomplete because information on dark pool and ELP fees and rebates are generally not readily available. But if brokers are now forced to collect this information and report it in aggregate for the 606 reports, brokers may be less reluctant to make these data available to clients directly. But even if brokers remain reluctant to share this information, buyside desks could use their 606 reports to approximate these fees/rebates for each broker on their own.
I also think that the SEC’s focus on execution practices indirectly reinforces the importance of buyside firms not only receiving the information they need to execute more efficiently, but also acting on that information. More broadly, as the pendulum swings in favor of the buyside in terms of the availability of execution information, the buyside is better able to optimize its trading and improve performance. But this also means that buyside firms can no longer hide behind the excuse that the data aren’t available to them. The data are out there, and more is coming – both from the SEC-mandated reports as well as from brokers providing more and more information on request. It is now up to the buyside trading desks to execute (bad pun intended, sadly).
In sum, the modification to Rule 606 not only add to the information set available to trading desk. They could indirectly result in even greater information being made available to clients via other means. In this way, the modifications will make Rule 606 reports more than simply canaries for coal mines -- and hopefully much, much more.
The author is the Founder and President of The Bacidore Group, LLC. For more information on how the Bacidore Group can help improve trading performance as well as measure that performance, please feel free to contact us at info@bacidore.com or via our webpage www.bacidore.com.
For an overview of our other services, please click HERE.
And please check out our other blog posts available HERE.
Copyright 2019, The Bacidore Group, LLC. All Rights Reserved.
[1] Technically, orders are categorized based on pricing relative to midpoint, i.e., less favorable, more favorable, or at the midpoint, as opposed to taking or providing liquidity.
[3] See “Can Brokers Have It All? On the Relation between Make‐Take Fees and Limit Order Execution Quality” in the Journal of Finance, Volume 71, pp. 2193-2238.
Kommentare