SEC Approves Yet Another Conflicted Order Type
We were very disappointed to read yesterday that the SEC approved Nasdaq’ Extended Life Priority Order proposal. As you may recall, we wrote two comment letters to the SEC urging them to reject Nasdaq’s proposal. Our main concern with Nasdaq’s ELO order was information leakage. We detailed this in our December 19, 2016 SEC comment letter:
“In return for giving retail orders queue position, Nasdaq will require that these orders be marked as extended life orders. This identifier will be disseminated to customers who purchase a proprietary data feed from Nasdaq. In other words, ELO orders will easily be identified by high frequency traders since they are the ones who most commonly purchase the data feeds. Knowing which orders in the queue that will not be cancelled or revised for at least one second and knowing that these are retail orders are extremely valuable pieces of information for a high frequency trader.”
The SEC did not think that information leakage would be a problem even though they acknowledged that Nasdaq would be informing consumers of their proprietary data feeds that a retail order existed:
“Orders marked with the new identifier—whether on an order-by-order basis or via a designated port—would be disseminated via Nasdaq’s TotalView ITCH data feed.”
The SEC apparently bought Nasdaq’s argument that the dissemination of this information is “transparency” rather than information leakage. We probably shouldn’t be surprised by this since the SEC also allowed Nasdaq to leak information for seven years about hidden orders that interacted with post-only orders.
Here is how the SEC justified the ELO information leakage:
“The Commission believes that market participants (retail and non-retail) are not likely to be detrimentally affected by other market participants’ knowledge, via the ELO identifier, that certain orders originated from retail investors and must remain unchanged for at least one second. In particular, information leakage would likely not be a concern for retail interest because retail interest is most often represented by one order at a single price. Also, the lack of an ELO attribute on any particular order would likely not allow market participants to say with any assurance that the order is of a particular participant type. Moreover, the Commission does not believe that identification of ELO orders would necessarily result in market participants choosing to route to ELO orders last and therefore result in lower fill rates for these orders. In addition, the Commission notes that the use of the ELO attribute is voluntary.”
Let’s break down each one of the reasons why the SEC thinks the information leakage from the ELO order would not be a problem:
- “Information leakage would likely not be a concern for retail interest because retail interest is most often represented by one order at a single price” – This is exactly the problem. The ELO identifier is essentially a flashing red light signaling that a retail order exists. This also means that if the order is retail, then it is not institutional.
- “The lack of an ELO attribute on any particular order would likely not allow market participants to say with any assurance that the order is of a particular participant type” – True. But as we know, the high frequency trading game is all about tilting the odds. Every bit of information in the proprietary data feeds gets analyzed and is then acted upon. We anticipate that patterns will form once retail brokers start using the ELO attribute and predatory HFT’s will be able to use this information to assist in their strategies.
- “The Commission notes that the use of the ELO attribute is voluntary” – Yes, we know ELO is voluntary. But so is the “RTAL” indicator that BATS EDGX uses to identify retail orders in their data feed. To entice retail brokers to use the RTAL indicator, BATS EDGX offers an enhanced rebate of 34 mils per share. Based on the retail broker 606 reports that we have reviewed, EDGX is the clear winner for attracting retail limit orders since they pay the highest rebate. While the SEC is correct that ELO is voluntary, we have a suspicion that Nasdaq will be offering an enhanced rebate to attract retail broker order flow. If they do this, then it tells us they are willing to lose money on the trade to ensure that their proprietary data feeds continue to include valuable information for the HFT community.
The bottom line here is that once again conflicted order types and information leakage from proprietary data feeds seems to be going unnoticed by the SEC. We tried our best to explain this to the SEC with our two letters but apparently Nasdaq was more convincing than us. Maybe Nasdaq argued that ELO orders could bring back some liquidity to lit exchanges that is currently being siphoned off by internalizers? We think more displayed liquidity is a good thing but not at the cost that Nasdaq is asking the market to pay.