This commentary is terribly late, but better late than never! I’ve had the links in my “Draft Posts” pile for a long time, but am only now getting around to posting the stuff.
The Canadian Securities Administrators (which is the securities commissions) and the Investment Industry Regulatory Organization of Canada have released POSITION PAPER 23-405 DARK LIQUIDITY IN THE CANADIAN MARKET, which provides an update to 23-308 and 23-404, which have been discussed on PrefBlog. The
One conclusion is:
In our view:
- An exemption to the pre-trade transparency requirements should only be available when an order meets or exceeds a minimum size (in the Position Paper, we will refer to this as the “minimum size exemption” or “minimum size threshold”). This minimum size threshold for posting passive Dark Orders would apply to all marketplaces (whether transparent or a Dark Pool) regardless of the method of trade matching (including continuous auction, call or negotiation systems), and for all orders whether client, non-client or principal.
They further explain:
Rule 6.3 of UMIR (the Order Exposure Rule) states that “A participant shall immediately enter on a marketplace that displays orders … a client order to purchase or sell 50 standard trading units or less of a security ….”. Aside from the specific exemptions under the Order Exposure Rule, it is currently required that client orders with a quantity equal to or less than 50 standard trading units will be directed to a transparent marketplace in order to be displayed. The Order Exposure Rule encourages transparency and supports the price discovery process, while still providing an opportunity for dealers to minimize large, passive order information leakage. Price discovery is enhanced by requiring smaller passive orders to be posted in a visible marketplace and rewarding those orders with increased execution opportunities. Additionally, IIROC has provided guidance in Market Integrity Notice 2007-019 with respect to the entry of client orders on non-transparent markets or facilities
So in other words, if you want to keep your order for 4900 shares dark – you can’t. It’s illegal. The powers that be have determined that your order is required as a critical part of the price discovery process. Those prefs you want to sell are quoted at 20.50-00, and you’d like to place an offer at 20.90, keeping it dark so the penny algos won’t move to 20.89? Tough luck, sucker. You are required to tip your hand to your broker, the exchange and the world.
The regulators don’t have a clue about the real world, though:
the Order Exposure Rule which requires that participants immediately enter on a marketplace that displays orders, all client orders for 50 standard trading units or less, subject to a number of exceptions. This is a benefit gained by passive, displayed orders in a transparent order book, in that active orders not meeting the size conditions of the rule are obligated to be routed to a transparent market, thus increasing the chances of execution for the displayed order.
If this was really beneficial to retail they wouldn’t need to make it mandatory. It is plainly apparent that not a single writer of this report has ever attempted to execute a trade in an illiquid market. They state that their motivation is:
The posting of limit orders in a visible book is important to maintain the quality of price discovery. To achieve this, limit orders should ideally be directed to, and displayed in visible marketplaces in order to facilitate the price discovery process.
In other words, the purpose of limit orders is not to save some money. The purpose of limit orders is to “facilitate the price discovery process.”. At least, according to the regulators.
It is in the discussion of the rule that the regulatory contempt for retail shines through:
allow larger orders to be executed with decreased market impact costs. However, as the “market impact cost” rationale described above may be less relevant to small Dark Orders, a possible rationale for allowing smaller orders to be posted as Dark Orders and be exempted from pre-trade transparency requirements is that they offer price improvement over the NBBO. While small orders may provide some price improvement when posted as a Dark Order, the limited quantity diminishes the value of price improvement to all market participants when compared to the value, or net benefit, of having larger Dark Orders offering the same price improvement, as well as providing much greater amounts of liquidity to the market as a whole. Currently in Canada, there are Dark Pools or Dark Order types that offer as little as 10% price improvement over the NBBO. In the situation where the NBBO spread for a particular security is very small (for example, one penny), we question whether the price improvement provided by small non-transparent orders is sufficiently meaningful for contra-side participants.
So if you feel that 10% price improvement (which, I take it, is 10% of the quoted spread, not the quoted price) is a worthwhile thing to have in your pocket – tough! Your orders are not for your benefit, they are for the benefit of other marketplace participants.
This is made explicit in the official staff view:
It is our view that the potential negative impact on price discovery of a greater number of small orders being entered without pretrade transparency and the potential drain on visible liquidity outweighs the benefits of the possible price improvement that they may offer. While post-trade information contributes to the price discovery mechanism, pre-trade transparency is an important element. The risk of a significant erosion of the quality of that mechanism exists if a substantial number of small orders are posted in the dark. As regulators, part of our mandate is to foster fair and efficient capital markets. The requirements to post small orders to a visible market and facilitating price discovery are key components of fair and efficient capital markets.
Consequently, we are of the view that an exemption from the pre-trade transparency requirements should only be available for orders meeting the minimum size threshold.
The staff, by the way, are not people with any knowledge of, or interest in, the capital markets and have no idea of what is meant by the phrase “fair and efficient capital markets.” They’re 25-year old B-School grads and 30-year-old lawyers with majors in boxtickingology.
They asked for responses on or before January 10, 2011.
One response which attracted some press attention was a tearful wail from David Panko of TD Securities, who is distraught that he cannot compete unless he has privileged access to the markets:
In addition, the existence of rebates encourages sophisticated high frequency traders to “stack the quote” by bidding and offering in large size. While this is certainly a benefit for small investors with orders that can be filled “on the quote”, it also disincents traders from placing natural bids and offers in the market, as these new bids and offers would be behind a large volume of HFT orders In turn, this leads to a greater proportion of client orders filled actively by crossing the bid-ask spread. The result is a combination of worse average fill prices for the end clients (through more frequent crossing of the bid-ask spread) and much higher marketplace fees for the broker (through a higher active/passive ratio)
Mr. Panko either does not know, or does not care, about the distinction between “informed” and “uninformed” traders.
These are the two types of traders who execute actively. Basically, so the theory goes, there is a “market price” for a security, which is the mid-point of the bid-offer spread. Call that “P” and the spread “S”. There is also a “Value Price” for the issue, P’, which is the Platonic Ideal price in a perfectly efficient world in which all private information has been put to work. It is assumed that over the long run P = P’, but these can vary over the short-run.
An uninformed trader will execute his active order and, on average, lose 0.5*S on each of his trades. If, however, an informed trader oberves that P’ > P + 0.5*S … back up the truck! He’ll execute all kind of active orders, happily paying the spread in exchange for getting his buys filled at a price that may be significantly less than P’.
So while one can certainly say that end-clients, as a group, are getting worse average fill prices, it is trivial to demonstrate that informed traders are getting better average fill prices and uninformed traders are getting worse ones. Informed traders will receive market rewards that reflect the accuracy of their estimates of P’.
To which I say – “Wonderful”. If there is anybody out there who truly wants to encourage price discovery as a valuable social good, they will do everything they can to reward informed traders and punish the uninformed. If we’re lucky, the uninformed ones will go bankrupt, go on welfare and die – that would be good.
Sufficient bankruptcy of the uninformed will, eventually, lead to a higher proportion of informed traders in the market, greater losses by HFT, therefore somewhat wider spreads, therefore more incentive to place “natural” bids and offers in the marketplace and give Mr. Panko something to talk about with his clients. It’s called competition.
So, with the shift to HFT stacking the quote, informed traders are winners – since they can execute at better prices. Uninformed active traders are also winners – since their market orders are executed at better prices. The only losers are the uninformed “market-maker” traders – those who, when attempting to buy a stock, consider it to be the height of genius to place a bid inside the quote rather than lifting the offer. These guys are now having their lunch eaten by people who can do it better. Boo-hoo-hoo. It’s called competition.
Mr. Panko also squares his rot for a good boo-hoo-hoo about how unfair it is for brokers to pay so much money for execution of active orders. Well, Mr. Panko of TD Securities, today is your lucky day because after a lifetime of research into market microstructure I think I’ve hit on a solution for you, which I will provide to TD Securites free, gratis and for nothing: if it costs more, charge more. A crazy idea, I know, but it just might work! Why should an off-quote limit order attract the same commission as a market order?
It’s a shame I’ve been so critical of Mr. Panko’s commentary, because his discussion was quite interesting, with more supporting numbers than is usual with self-serving sell-side lobbying letters in Canada, although the numbers themselves did not get much support. Unfortunately, the decision to “strongly recommend” the complete elimination of liquidity rebates is not particularly well supported by his discussion and argumentation: perhaps TD Securities wrote the recommendations in the executive suite, then assigned him the task of writing the rest of the letter.