HFT Myths
- Thread starter hft
- Start date
Another ignoramus curious about "offset order routines".
Thanks
Thanks
I agree with most of your sentiments, but IEX has "broker preferencing" (queue jumping for the "right participants"), if IEX is approved as an exchange with that feature, it will be the most unfair exchange of all. And then, what's to stop every other venue from trying to get SEC approval for queue jumping?
Moreover, according to the Lewis book that painted such a great picture of IEX, it's not the brokers that really get to take advantage of this, but rather HFT's who buy that flow (see the footnote on the bottom of page 54 of the Lewis book) -- no doubt the same oligopoly of HFT's that are already entering into payment-for-order-flow deals with US retail brokers to keep orders off of the lit exchanges. Who loses, in the end? Basically everyone, except the brokers (small winners), and a very small group of big HFT's (big winners).
tabbforum.com/opinions/iex-broker-preferencing-and-the-queue-jumpers
Moreover, according to the Lewis book that painted such a great picture of IEX, it's not the brokers that really get to take advantage of this, but rather HFT's who buy that flow (see the footnote on the bottom of page 54 of the Lewis book) -- no doubt the same oligopoly of HFT's that are already entering into payment-for-order-flow deals with US retail brokers to keep orders off of the lit exchanges. Who loses, in the end? Basically everyone, except the brokers (small winners), and a very small group of big HFT's (big winners).
tabbforum.com/opinions/iex-broker-preferencing-and-the-queue-jumpers
that is a very contentious argument and there is simply not enough evidence to conclude whether the net effect will be positive or negative. But I agree that the perception of unfair markets may be raised.
Are they doing broker preference like they do in Toronto? Where orders that would remove liq first check if they can remove an order from the same broker and then the general queue?
(Tried to read the tabb link but it seems broken)
Let me try the link again -- it seems to have been transformed oddly in the earlier version -- maybe something to do with the ET upgrade:
http://tabbforum.com/opinions/iex-broker-preferencing-and-the-queue-jumpers
As per the Lewis book reference, it's on page 54 of Flash Boys. That's the part that claims that it's really the HFT's who are really taking advantage of this, by paying e.g. CIBC or RBC for their "broker preference". So it's really HFT's who get to jump queues -- if they're big enough and willing to pay for the "relationships" with the brokers. Result: hampered price discovery and a market that's unfair to everyone but an oligopoly of HFT's who can milk it for all it's worth without the "headache" of truly competitive pricing.
http://tabbforum.com/opinions/iex-broker-preferencing-and-the-queue-jumpers
As per the Lewis book reference, it's on page 54 of Flash Boys. That's the part that claims that it's really the HFT's who are really taking advantage of this, by paying e.g. CIBC or RBC for their "broker preference". So it's really HFT's who get to jump queues -- if they're big enough and willing to pay for the "relationships" with the brokers. Result: hampered price discovery and a market that's unfair to everyone but an oligopoly of HFT's who can milk it for all it's worth without the "headache" of truly competitive pricing.
Last edited:
thnx 4 reposting. great article.
I traded TSX for a few years and was pretty familiar with its matching engine up to a couple of years ago. What they say about the way preferencing pushes people to trade with the largest brokers is very true. At one point we switched from a very small broker to a mayor bank and the difference in order flow was quite significant.
Although I think that it doesn't favor the HFT crowd as much as US style internalization, since any trader can place an order into the queue and the differences between one queue and the other have to do with your choice of broker, not with your ability to decide whether or not you want to fill an order in less than 100 milli.
I admit I'm a bit of a wally with regard to equity market microstructure, but this my understanding of the broker priority (i.e. preferencing) issue. Please explain where I am wrong. Flames and insults will be ignored unless they contain information.
I think Katsuyama would say that the effect on the customer is neutral. It works like this:
--- 1) Broker gets a limit buy order from customer A and posts it at the end of the queue
--- 2) Same broker then gets a limit sell order for the same price from customer B
--- 3) IEX crosses A with B, thereby allowing both orders to jump to the head of their queues
If the cross of A and B is not done on the exchange, the broker will simply cancel order A and cross A and B on his dark pool. It makes no price or time difference to the two customers whether the trade is done on the exchange or the dark pool. To use Tabb's metaphor, the customers can jump the theater queue or enter by a special door. The broker benefits by internalizating the trade. The equities market benefits in general by moving the trade from the dark pool to the lit exchange for greater transparency.
The article acknowledges that preferencing is a wash that moves some amount of off-exchange internalization onto lit venues, but goes on to express concern that the tradeoff is a more complicated and unfair market place. That is a reasonable concern, but the author doesn't spell out how that works. Most of the author's objections seem to stem from internalization. I agree that internalization is an issue that should be address by regulators. I also agree that regulators should take a hard look at whether broker priority creates unanticipated consequences. So just what are those consequences?
IEX states that members of an IEX exchange must be broker/dealers. So HFTs can only benefit if they are broker/dealers, and being a broker/dealer allows them to internalize their orders. What difference does it make whether they internalize orders on a lit exchange or a dark pool?
Here are links to Katsuyama's response to the issue:
video: http://www.stockboard.com/blog/view.php/IEX-president-Katsuyama-addresses-broker-priority
transcript: http://www.cnbc.com/id/101715469
I admit I'm a bit of a wally with regard to equity market microstructure, but this my understanding of the broker priority (i.e. preferencing) issue. Please explain where I am wrong. Flames and insults will be ignored unless they contain information.
I think Katsuyama would say that the effect on the customer is neutral. It works like this:
--- 1) Broker gets a limit buy order from customer A and posts it at the end of the queue
--- 2) Same broker then gets a limit sell order for the same price from customer B
--- 3) IEX crosses A with B, thereby allowing both orders to jump to the head of their queues
If the cross of A and B is not done on the exchange, the broker will simply cancel order A and cross A and B on his dark pool. It makes no price or time difference to the two customers whether the trade is done on the exchange or the dark pool. To use Tabb's metaphor, the customers can jump the theater queue or enter by a special door. The broker benefits by internalizating the trade. The equities market benefits in general by moving the trade from the dark pool to the lit exchange for greater transparency.
The article acknowledges that preferencing is a wash that moves some amount of off-exchange internalization onto lit venues, but goes on to express concern that the tradeoff is a more complicated and unfair market place. That is a reasonable concern, but the author doesn't spell out how that works. Most of the author's objections seem to stem from internalization. I agree that internalization is an issue that should be address by regulators. I also agree that regulators should take a hard look at whether broker priority creates unanticipated consequences. So just what are those consequences?
IEX states that members of an IEX exchange must be broker/dealers. So HFTs can only benefit if they are broker/dealers, and being a broker/dealer allows them to internalize their orders. What difference does it make whether they internalize orders on a lit exchange or a dark pool?
Here are links to Katsuyama's response to the issue:
video: http://www.stockboard.com/blog/view.php/IEX-president-Katsuyama-addresses-broker-priority
transcript: http://www.cnbc.com/id/101715469
IEX's 'small market share' excuse for queue jumping is absurd; what's to keep other exchanges from this model if IEX gets it approved? In fact, other exchanges have tried for this in years past, and the SEC has rightly rejected them. Now, IEX may be able to force queue jumping through the SEC on the back of Lewis's sensationalistic yet naïve book. Broker preferencing yields no benefit to the end investor or companies wanting to raise more capital -- it's just a hidden tax on everyone except certain brokers (those with undirected customer flow to skim, which includes virtually all "discount" retail orders) and their HFT/wholesaler "affiliates".
'Broker preferencing' would also create a back door in case Congress (very rightly) decides to act against the payment-for-order-flow schemes now used by the oligopoly of big wholesalers (such as Knight), so the wholesalers can continue their queue-jumping rackets without needing to use PFOF as they do presently. (See footnote, Flash boys, page 54, for an interesting description of the situation in Canada, which is a lot nastier and less transparent than it first appears. )
I think Katsuyama's excuses for including queue jumping in IEX are inconsistent and ultimately hollow -- why would IEX put something so grossly unfair in a venue that otherwise is supposedly designed to make the markets more fair? This makes it seem to me that IEX never really had much to do with fairness, and that its struggle to succeed by whatever means necessary will further fragment the already-messy US equities system in a manner that will further entrench an oligopoly of HFT's.