SEC Steers Pilot on Equity Trading Fees and Rebates Amid Headwinds
Diving into one of the more controversial issues in equity market structure, U.S. regulators are moving forward with the transaction-fee pilot that will force stock exchanges to lower the fee caps on rebates they pay to brokers and market makers.
I also thought the comments from Sal Arnuk and Joe Saluzzi at Themis Trading were very interesting:
Themis Trading appreciates the opportunity to comment in support of the Transaction Fee Pilot for National Market System Stocks proposed by the Commission.
We are an institutional agency broker that trades in US equity markets on behalf of our clients, who are institutional firms that represent millions of regular investors. Our market structure viewpoints are based solely on our interactions and experiences trading in dozens of exchanges and dark pools in the US. We have no proprietary trading. We operate no trading venue, nor do we have any ownership stakes in any trading venue – lit or dark. Our only interest is the best trade execution possible for our buy side clients. It has been our only interest since our inception in 2002.
We will spare you a recounting of Modern Market Structure History; instead, we simply direct you to two places:
1) Our 2012 book, Broken Markets, and specifically Chapter Thirteen, where we recommended the elimination of the maker-taker pricing system. Our belief today is the same as our belief then, which is that payment for order flow is a conflict of interest that should be banished from our market structure in all its forms, including maker-taker.
2) The testimony of Joe Saluzzi before the U.S. House of Representatives Committee on Financial Services on June 27th 2017 (link here). Specifically, we would appreciate your reading of two sections: Stock Exchange Rebates, and Payment For Order Flow and Order Routing.
II. Transaction Fee Pilot For NMS Stocks
We commend the Commission for undertaking this pilot program. It is clear to us that Mr. Redfearn and his team has given a great deal of thought in crafting it. It is magnificently well-written, thorough, and clear.
II. Our Comments
In past debate where we have advocated for eliminating the conflicts of interest caused by payment for order flow in general, including the Maker-Taker rebate pricing system, based on Common Sense, we were rebuked for not basing our thoughts and opinions on this issue on concrete data.
Tired of the Edwards Deming quote “In God we trust; all others bring data”, we were pleased and cautiously optimistic when NASDAQ sought to undertake a study in 2015 to test whether higher access fees were discouraging use of the public markets. Yet even then we noted that the NASDAQ experiment would not yield terribly useful results. After all, it was conducted unilaterally on just one exchange of over a dozen, in a limited subset of stocks – 14 to be exact, and only over a four month period. SIFMA also made this point in their January 30th 2015 Comment Letter on the NASDAQ Proposal:
“the proposal’s limited scope and application cannot act as a substitute for a market-wide access fee reduction that would change the dynamics of access fees and rebates across the entire market. For the proposal to accurately measure the structural impact of reduced access fees, the proposal should be carried out across all exchanges and with a larger sampling of symbols.”
It has been clear to us, and many others, that a more comprehensive multilateral market-wide approach would be needed to yield usable data that could be used to test how lower access fees, and a lack of rebates, would impact market quality and marketplace behavior.
Over a year later, in April 2016, the EMSAC Regulation NMS Subcommittee delivered its Memorandum for a Framework For Potential Access Fee Pilot. It did not include a “no-rebate” bucket; EMSAC could not achieve consensus on two issues: the prohibition of rebates, and the inclusion of inverted venues in the Pilot.
Now, nearly six years after our book Broken Markets was published, nearly three years after the NASDAQ experiment in 14 stocks, and nearly two years after the EMSAC Memorandum, under the new Trading and Markets leadership of Brett Redfearn, we are quite pleased to see the SEC Transaction Fee Pilot laid out in such a well-designed way:
1) We are pleased that the Transaction Fee Pilot includes all NMS exchanges, including inverted exchanges. For a true testing of marketplace behavior, as we stated in our March 9th 2015 blog post, “the SEC should take a leadership role, and order this experiment market-wide for all of the exchanges. If the SEC did that, then we really would have a truly important set of data with which we could all assess effects on modern market quality”.
2) We are pleased that it will include 1,000 stocks in each bucket. We do believe that the stock selection for each bucket should span multiple market caps.
3) We are pleased that the Commission included a “no rebate” bucket, and that it also was thoughtful in prohibiting “linked pricing”. The Commission is demonstrating great care in attempting to keep conflicts of interest from Test Group 3.
4) Test Group 3 does allow exchanges to offer non-rebate based linked pricing for registered market makers tied to new market making standards. To our knowledge, well-capitalized registered market making firms (such as Virtu) are not opposed to market making standards. This seems a sensible exemption, provided that any new market making standards are meaningful, and monitored closely by the exchanges. While we would not have proposed such an exemption, we understand the challenge the Commission faces in getting buy-in from our Modern Markets. This compromise, if monitored closely, demonstrates the Commission’s flexibility.
5) The Transaction Fee Pilot does not address ATS’s /off-exchange liquidity-provider ping pools. While the reason given is the different regulatory regime that exists for those revenues, we want the Commission to understand that often the market makers on the exchanges are also involved in off-exchange market making. We are concerned that Smart Order Routers (SORs), especially with stocks in Test Group 3, will increase their sending of investor orders to market makers off-exchange for enticements. These enticements might be in the form of a free fill, a rebated fill, or even simply the economic clout tied to a large prime broker relationship. We caution the SEC to watch for these interactions, as they certainly can increase investor costs.
We caution the SEC to understand that the calculation of these costs is also not simply tied to some comparison price of each child order versus the market at the time of the fill. After all, the price at the time of the received route is often different than the time of the received fill. More importantly, market makers will be modeling ongoing participation of long-running orders, and trading alongside them for periods of time; this introduces even more substantial implicit cost.
Section 11a of the Securities Exchange Act states on page 104 that it is in the public interest, and appropriate for the protection of investors and the maintenance of fair and orderly markets, to assure (among other things) an opportunity for investor orders to be executed without the participation of a dealer. Even back then the Commission understood that with intermediation comes investor cost.
There are two truths that we have learned at the beginning of our careers nearly thirty years ago that still hold today:
1) The cheapest way for investors to trade is naturally. By this we mean without intermediation.
2) “It’s easy to make money trading every day when you have the tickets in your pocket”.
Our modern markets have evolved, not by accident, to maximize the insertion of intermediaries between natural buyers and sellers. Stock exchanges that employ maker-taker pricing sell speed in ways designed to attract the fastest hyper-efficient insertion of these intermediaries between natural buyers and sellers. This Transaction Fee Pilot provides an opportunity, specifically in Test Group 3, to test whether or not this excessive intermediation is good for investors, and we commend the Commission for undertaking it. We believe that forcing out some of the conflicts in order routing will improve market quality and lower investor costs.
We expect that the majority of commenters from the institutional trading community will support the transaction fee pilot. These institutional investors have grown weary of the maker/taker model and no longer want their orders subject to the economic incentives of their broker.
However, we expect that there will be also be a large community of dissenters. This group will most likely consist of the three major exchanges (NYSE, Nasdaq and CBOE) as well as a number of market makers. Their arguments will be based on their own profit incentives and not what is best for the market. They will likely use one or more of the following arguments to try and convince the SEC not to undertake a pilot:
1) If institutional clients are unhappy with the routing of their orders, then they should fire their brokers.
This is not as easy as it sounds. Many institutions are still tied to large broker/dealers because of research commitments and other services that the broker provides. They can’t just fire their brokers.
2) Exchanges will raise other fees and add new exchange medallions to offset the lost revenue from lower access fees and potentially less volume.
This is an idle threat since the SEC has the power to reject any new exchange application as well as not approve new fees that are excessive and unwarranted.
3) Spreads will widen and liquidity will decrease.
There is no evidence of this. In fact, rebates of 1/3 of a penny represent only a small fraction of the spread in many securities that are not in the top 100 volume. The purpose of the pilot is to test this hypothesis with actual data. Before exchanges and market makers make this argument, they should wait to see the data.
4) Exchanges will threaten to sue the SEC for this “draconian” proposal.
Another idle threat that we doubt would ever go far. Exchanges should worry about their own legal problem right now – the possibility of losing absolute immunity – rather than threatening the SEC.
5) The market should dictate the transaction fee cap.
An arbitrary 30 mil fee cap already exists. The transaction fee pilot will test other potential caps for two test buckets but will let also let the market decide the access fee in the third no-rebate bucket.
6) Before conducting a pilot, the SEC should take action on two outstanding proposals: disclosure of order handling information and the amendments to Reg ATS.
The SEC has already indicated that they are moving on these proposals and that we should expect some action on them in the coming months, likely before the transaction fee pilot were to begin.
7) Don't undertake a pilot but rather do a "holistic review" of equity market structure.
This is the standard argument that the industry has been using to slow down market structure reform for the past decade. They do not want any change to the very profitable model that they have built since Reg NMS took effect in 2007. The entire industry sees through this argument; hopefully the Commission does as well.