Canadian securities regulators are finally dipping their collective toes into the murky waters of “dark” trading by proposing some limits on this emerging corner of the securities market.

Equities market structures have been rapidly evolving in many parts of the world over the past couple of years, often outstripping regulatory regimes that had been designed for traditional exchange-floor trading. In the autumn of 2009, the Canadian Securities Ad-ministrators and the Invest-ment Industry Regulatory Or-ganization of Canada had published a joint consultation paper on one prominent aspect of that evolution, the emergence of so-called “dark liquidity” — trading in dark pools, or the use of dark orders, that don’t provide pre-trade transparency. And the regulators followed that paper with a public forum on the issue last spring. Now, they are getting around to setting some policy in the area.

In a paper jointly published last month by the CSA and IIROC, the regulators are taking a position on some of the complicated market-structure issues that they have been thrashing around for the past year. In doing so, the regulators are explicitly recognizing the benefits of trading in the “dark”; but they also are starting to set some limits on the practice.

The regulators explain in their report that while they generally believe that trading activity should be transparent, they also recognize that there are benefits to secrecy, both for the individual trader and for the market as a whole.

Dark pools initially had developed to allow large orders to trade anonymously, minimizing the market impact and costs that a large, visible order could incur. Proponents of dark trading also argue that this process can increase overall market liquidity by having dark orders available on a public market rather than sequestered in the “upstairs” market.

On the other side of the discussion, critics of dark trading have argued that the process is inherently unfair, because it allows traders in these venues a free ride on the pricing information supplied by the traditional, “lit” markets. Transparency should be paramount, these critics maintain. They also suggest that dark trading poses a fundamental risk to market integrity, undermining the traditional markets by drawing liquidity away from them.

The CSA/IIROC report acknowledges these concerns, admitting that if orders that would traditionally be sent to visible marketplaces end up trading in the dark, “…there could be a negative impact on the price discovery process and the liquidity available.”

That possibility is increasingly a worry for regulators, as the nature of dark trading morphs from being just a way for institutional investors to get large orders processed and into a trading tactic that is used for small orders as well. Indeed, the regulators note that dark trading is now being used to do more than just execute large trades without moving the market; it also is being used to protect proprietary trading information and hide from algorithms seeking out that sort of information. Traders may use dark trading simply to lower trading fees or to seek out better pricing. And, in some countries, the CSA/IIROC report notes, brokers are developing dark “pools” to internalize order flow.

As dark trading takes on greater prominence, the regulators also are concerned that the risks the process poses to market integrity are growing. Thus, they are proposing to set some parameters on the practice.@page_break@Under the proposals, only orders that meet a minimum size threshold are to be allowed to trade in the dark (and, thus, be exempt from pre-trade transparency requirements). The regulators’ rationale for this position is that it respects the original purpose of dark pools (trading large orders anonymously) but avoids some of the negative consequences that may arise as dark pools stray from that function.

Says the CSA/IIROC report: “The risk of a significant erosion of the quality of [the price discovery] mechanism exists if a substantial number of small orders are posted in the dark. The requirements to post small orders to a visible market and facilitating price discovery are key components of fair and efficient capital markets.”

Ultimately, the report concludes, the “potential negative impact” on price discovery and the drain on visible liquidity that would occur if more and more small orders are entered without pre-trade transparency outweighs the likely benefits. The regulators’ aim is to allow large orders to trade in the dark while protecting the traditional, visible market from too much migration to the dark by small orders.

Yet, while the regulators have decided that there should be a minimum threshold before an order can be placed in the dark, they haven’t decided what that cut-off should be. That’s one of the questions they now are seeking comment on. (Comments are due on Jan. 10, 2011.)

The regulators have decided that the proposed minimum threshold will apply to both dark pools and dark orders, regardless of the method of trade matching or order type. Traders will not be allowed to aggregate orders to meet the threshold, and cannot reduce an order to a level below the threshold once it’s posted (although partially filled orders that subsequently fall below the minimum threshold will be allowed to remain in the dark until they are executed or cancelled).

The CSA/IIROC report also recommends allowing two dark orders that meet the minimum size threshold to be executed at the best available price (the national best bid and offer), and to be executed ahead of visible orders. The report notes that while the regulators agree with the general consensus that visible orders should be given priority over dark orders at the same price, they also believe that the benefits of dark trading are substantial enough to justify allowing these orders to jump the queue when it involves two large, dark orders.

The report adds that while the regulators want to encourage displayed orders, they also believe that the benefits of dark trading are valuable enough that the regulators don’t want to penalize genuine, large dark orders by requiring price improvement in every situation. However, dark orders will have to provide “meaningful price improvement” — which the report defines as at least one tick; in cases in which the bid/ask spread is already at the minimum tick, meaningful improvement would be at the midpoint of the spread.

This definition prevents trading from occurring in ever-smaller increments, ostensibly to provide vanishingly small price improvements but really just to get executions ahead of visible orders. IE