You have been granted access to this page through First Click Free. Subsequent use of TabbFORUM will require logging in. If you don't have an account, registration is free.


  • Rail_thumb_craig_beddis-automic-slaying_complexity_through_process_automation

    Slaying Complexity With Process Automation

    Technical fragmentation within capital markets firms has led to tremendous complexity, which in turn is driving up operational risk and cost. According to Craig Beddis, managing director, Automic, the depth and ...
  • Rail_thumb_dan_schleifer-chartiq-redefining_financial_visualization

    Redefining Financial Visualization

    With the ongoing consumerization of IT, we now expect to be able to access everything – from both our personal and professional lives – from a mobile device. The one function in the capital markets that ...
  • Rail_thumb_mahesh_muthu-eclerx-the_power_of_data_to_ease_collateral_mgmt

    Unlocking Data to Power Collateral Optimization

    The ISDA Master Agreement and Credit Support Annex (CSA) documents govern the relationships between counterparties to an uncleared derivatives trade. And the information locked in these documents is critical ...

More Video | Podcasts

Enabling Platforms For Fast Markets

19 April 2013

Biggest Dark Pool Goes Silent

Every month since March 2007, a wide swath of dark pool operators have voluntarily disclosed various execution statistics. Now these reports are in the danger of becoming inaccurate, as Credit Suisse has decided to stop reporting.

Credit Suisse, the behemoth of US institutional equity electronic trading, has decided to no longer publish the volume statistics of its Crossfinder dark pool.  As the publisher of the LiquidityMatrix, TABB Group has a vested interest in seeing dark pools voluntarily submit their execution statistics to us; however, I will try and address this as an objective observer. In the past there typically have been four main reasons why a dark pool would not submit its volume figures to us:

  • A lack of volume or a persistently negative trend in volume.
  • A concern that the disclosure would reveal too much about the firm’s operations or client behavior.
  • A disagreement over the methodology other dark pools use to count volume (typically this comes down to claiming that much of what happens in some pools is merely a form of routing).
  •  The company (or division) is inherently secretive and doesn’t want to disclose anything.

All of the 12 dark pools that have not reported to us fit into one or more of the categories above. 

Note: ITG reports quarterly.
Source: TABB Group

To understand why Credit Suisse has stopped reporting, we have to also consider why any of the 15 dark pools that have reported execution statistics ever bothered. First, dark pools never submitted their trading statistics for purely altruistic purposes. The LiquidityMatrix is a league table representing the ability of firms to match customer orders in an anonymous fashion at a discounted price. The importance of publicizing numbers has grown over the years. There is correlation between the top dark pools and the top algorithmic providers, so the LiquidityMatrix became a way to help sell execution services.

But as the controversy over dark trading has erupted over the past year, fueled by the rise in off-exchange trading in the US and the changes occurring in Canada and Australia, the industry has focused more on the gross amount of shares traded off-exchange, rather than the performance of any specific dark pool or other alternative trading systems. And, of course, there is the most recent spat between Credit Suisse and NYSE on the definition of dark and off-exchange volume.

At some point, you need to take a step back and ask whether all of this visibility is good for the rest of the company. What is it actually accomplishing? When you realize “going out loud” is only getting you burned, you go dark.

[Related:Applauding Dark Pool Regulation”]

A number of dark pools have been self-reporting these numbers to us since March 2007. The reporting has helped the industry understand how the behavior of market participants changes during periods of extreme volatility and low liquidity. The numbers also help the industry understand the changing demographics of the market. It is a darn shame that the industry will now lose an important component of that measure.

We also expect that the other major dark pools will reconsider their own reporting strategies. For some firms, the absence of Credit Suisse poses a real opportunity to capitalize on recent gains (see interactive chart). This event will allow Credit Suisse’s competitors to engage with clients on a topic that has become a bit worn. However, we also expect that other banks will take this opportunity to pull back in hopes that the industry publications will find something else to write about instead of the monthly change in on-exchange and off-exchange volume.

From our perspective, we will continue to take the sum of the reporting dark pools and estimate the total size and market share of dark pool venues, just as we have for years. Frankly, our estimate will not be as accurate as it was in the past, because we will be missing the largest dark pool. But we still feel it is important for the industry to continue to try to measure these numbers. At least until regulators feel the same and put in place a better way for the industry to measure the volume and impact of different trading mechanisms. We are not holding our breath.

Spotlight-white-trans For more stories in the Enabling Platforms For Fast Markets Spotlight Series click here.

Comments | Post a Comment

4 Comments to "Biggest Dark Pool Goes Silent":
  • Comment_main-thumb-3378195-200-eeqk8jvi8vsk38mxx85ssngsoii7elay

    19 April 2013

    I'll be perfectly honest with you Adam, the numbers are always met with a bit of skepticism and the way and how people report them is not regulated so there is no one to say whether they are accurate or not . Despite both you and Rosenblatt's best of intentions,  I can point out a number of inaccuracies in Tabb's Matrix alone.  You give "single counted" volume to dark pools that are clearly double counting.  At the end of the day I am not sure what difference it makes, but at least you have the most altruistic of intentions.  

  • Comment_steve-wunsch

    22 April 2013

    You could have added a fifth bullet, Adam, namely what might be called regulatory politics, which is better able than the others to answer the critical question: Why now?. It may not be a coincidence that the largest dark pool has pulled back its data just as the total amount of dark trading is nudging the percentages that some studies have claimed will materially disturb price discovery and thereby overall market quality. As an aside, I disagree with these claims. I don't believe any level of dark percentages is harmful to markets, as it merely represents the activity of size players finding consensus that the current price is OK, which is itself a form of or contributor to price discovery and, as all such activity is, is inherently stabilizing because it aids in the discovery of good prices. But the studies imply the opposite, and the regulators have their knickers in a bunch over the consequent opportunity to use these incorrect studies to further their transparency crusade by cracking down on dark pools. If they do lash out at dark pools, such as with a trade-at rule, that will only further HFT, as every move toward greater transparency has done, and will not improve market quality. But it is probably in fear of such regulatory over-reach that the big dark pools have started to again go fully dark. As a strategy, it probably won't work, and may backfire as it makes the regulators even more determined to take away their toy. But recognizing the suddenly hightened regulatory risk to their businesses caused by the combination of their aggregate market share percentages and these studies is probably the best place to go to start to understand their behavior. 

  • Comment_steve-wunsch

    22 April 2013

    I agree with your main conclusion, Adam, which implies that CS was aware of the potential for a regulatory backlash from withdrawing. But I think they may have gone ahead with their decision anyway, because they felt cornered. If they expected the backlash regardless of what they did (because of the studies and the hightened political criticisms of darkness, etc.), they may have felt they had little to lose by restating the original case for a completely dark pool, which more than most they seemed to hear demand for from their customers, and which they had eloquently defended in the past. And who knows, perhaps defending again the original rationale for their business may result in more market share and, if so, some help on the political front as others go dark too in order to chase that demand with them.

  • Comment_laurie_berke_s

    22 April 2013

    In developing dark order types, the exchanges joined the ranks of dark pool providers.  Adam is absolutely right that in the early days, advertising your ability to capture the lions' share of broker-sponsored dark order flow was a great method of advertising and attracting just that much more flow.  Today, CS has drawn a line in the sand.  You can't have it both ways.  If it is in the public interest to publish dark pool volume stats ( and the fact there is no standard is not the point here), then the exchanges themselves should do so.  If the exchanges deem that they would be harming their franchise in some way by so doing, then itI wouldn't be surprised to see everyone else come to the same conclusion in short order.

You must log in to comment.