Is There a Dark Pool Witch Hunt Going On?

As George Carlin once asked, “Is there another word for synonym?”

When it comes to the name “Dark Pools,” I wish we had followed our instincts – A pivotal moment for equity market structure occurred during a call of the old Securities Industry Association (SIA) in the early part of last decade. We were discussing the launch of several crossing engines as Alternative Trading Systems (ATSs) and brokers were branding them as a means of establishing their value. At the time, I distinctly remember suggesting that we should call them Quiet Pools instead of Dark Pools. Sadly, despite the general agreement that “quiet” was actually a more accurate description, there were other considerations that killed that idea; One large ATS had already launched an expensive marketing campaign around the name “Dark” and objected to a change. Someone else on the call said “What’s in a name? ” and the conversation ended…

Today, we know that what you name a product matters. Over the past several years there have been books, editorials and a wide variety of people condemning Dark Pools. There have been calls by legislators to “bring trading back into the sunlight” and some of the largest fines and most publicized regulatory actions have been against dark trading venues. This has all happened despite the important role that Dark Pools play in helping institutional traders manage transaction costs. When working large trades, it’s important to keep them quiet in order to minimize adverse price moves that can occur when proprietary traders learn that there are large buyers or sellers. Dark pools are a tool used by institutional traders to help control information leakage, but, despite that reality, the name conjures up images of nefarious behavior, which makes them an easy target for opponents.

If there was any doubt about the heightened concerns surrounding “dark” liquidity, you just need to look at the amount of discussion the issue received during the SEC market structure advisory board’s first meeting. That meeting was focused on Rule 611, the order protection rule, which explicitly makes trading on Dark pools harder, by preventing them from trading through lit venues posting a quotei. That fact did not stop various commenters from equating the “fragmentation” of our modern, competitive market that was promoted by 611, with the expansion of trading in the Dark. In fact, in the summary of the market advisory boards first meeting produced by Williams and Jensen for the STA, the word “Dark” appears 16 times… The fact is, however, the idea of matching buyers and sellers away from exchanges, is not new; it has always been one of the primary functions of broker-dealers.

Dark Pools critics often try to make a distinction between the function of broker dealers and the pools themselves. Critics rarely argue that broker dealers shouldn’t efficiently match buyers and sellers, yet Dark Pools are now integral to that function. In the last century (i.e., the 1990s), there were, literally, thousands of sales traders in New York, Chicago, Boston and San Francisco. Their main function was to either “work” orders, by finding the other side or solicit clients for orders to work. Despite the fact that the majority of those individuals have been replaced by a combination of trading algorithms, Dark Pools and the technical/quantitative staff who support that technology, the function remains the same. For brokers, matching their own buyers and sellers, is one of the most important services they can provide, whether provided by highly paid sales-traders or modern technology. The matching function cuts transaction costs for clients, by minimizing fees and reducing the bid offer spread clients pay, while reducing the amount of information leakage into the market.

Many firms built their first Dark Pools as a means to replace the old “hoot and holler” telephone and speaker systems that were used to help sales traders pair-off buy and sell orders. The old system was often referred to as the “upstairs” market and involved hundreds of sales-traders, at the large firms, calling investment managers to try to find the other side of large orders.

Unfortunately, for sales-traders, the expanded use of technology made their jobs more difficult. Algorithms sped up the trading process, which made it much harder to match buyers and sellers on the telephone.

Meanwhile, the algorithms, themselves, faced restrictions that made it virtually impossible for them to match buyers and sellers, since rules prohibit trading multiple orders together. Despite that limitation, algorithm usage grew and the amount of off-exchange trading dropped significantly. Of course, necessity being the mother of invention, electronic trading departments built crossing engines within their platforms to compensate. The algorithm software would continuously post small pieces of their orders in the brokers own Dark Pool and seek liquidity in the pool, before sending aggressive orders to lit exchanges. The byproduct was crossing rates improved in the aggregate, but the methodology resulted in many small executions in the dark, rather than fewer, but larger crosses with sales-traders crossing these orders on the telephone. Today, the order size being small is often cited as proof that institutional orders don’t really use Dark Pools. The truth is that it only signifies that the regulations make it difficult for those orders to be crossed in bulk.

Of course, none of this changes the fact that much of the activity in Dark Pools is not well understood by the investing public, nor by many professional investors. In particular, we have identified major deficiencies in both routing disclosures (Rule 606) and best execution reporting (Rule 605) and suggested improvements to both rules. Rule 606, which was supposed to provide transparency on order routing practices, needs a serious overhaul. There are two key changes that would help investors understand more about Dark Pools: First, 606 should include basic, aggregated best execution metrics for each main category of order. This would help investors discern if brokers using Dark Pools can demonstrate the benefits of such routing. Second, 606 should include full aggregated disclosure of all routing, executed or not, regardless of who operates the router. This would show, for the first time, the extent of order routing being done by brokers and exchanges, including all attempts to find liquidity in Dark Pools. Most importantly, these changes to rule 606 would provide a foundation for the development of more custom and detailed analysis for broker dealers to provide to their clients.

Rule 605, which is designed to provide best execution disclosures for all market centers, only provides valid metrics for retail market makers. While it has been a major success in driving down transaction costs for retail marketable orders, the exemption of “not held” and orders over 10,000 shares, has meant that it is, essentially, useless in evaluating anything else. If the SEC were to eliminate exemptions and implement our suggestions to modify rule 605, however, it would help investors understand the value (or lack thereof) of dark venues better. In particular, our recommendation that all immediate or cancel orders be grouped and analyzed separately, would facilitate the evaluation of the liquidity provided by dark pools.

The improvements we have suggested, would serve to dramatically improve the data available for evaluating the execution quality of the US Equity Market. Securing access to this data should be a prerequisite to structural change, especially since the SEC chair and other senior regulators, repeatedly, argue in favor of being data-driven.

Despite the negative publicity and fear of “the dark” we would argue that the most important change should be to Rules 606 and 605. Our proposals would provide a framework for comparing Exchanges and Dark Pools in a statistically valid way and help provide all investors access to the information necessary to evaluate all venues. This approach would help to “illuminate” the situation and provide actionable data for discussing potential structural change.

i The Order Protection Rule directly impacts the ability of venues to attract market makers to provide block liquidity outside of the bid ask spread. Under rule 611, a market maker willing to buy or sell much larger size than is displayed in the lit markets must accept the fact that the dark venue must broadcast that block to the market immediately. The rule requires that the dark pool send Intermarket Sweep Orders (ISOs) to all exchanges posting at superior prices and report the block as soon as the system can do so.

David Weisberger is the managing director and head of market structure analysis of RegOne Solutions.

The views represented in this commentary are those of its author and do not reflect the opinion of Traders or its staff. Traders welcomes reader feedback on this column and on all issues relevant to the institutional trading community. Please send your comments to