Sorry I have been absent from the MatchPoint blog for a few months. I have been recovering from a detached retina and it is hard to write when you cannot see…but there is, I suppose, something fitting in a dark pool provider having a spokesperson who cannot see. Though I am beginning to realize that what we may see before us can often prevent us from seeing the real truth. So let me get started….
For many months now, dark pools have been all the rage. All we have to do is look at the volumes to see that between 500 million and upwards of 1 billion shares a day trade through dark pools. (Lord knows how much is actually posted or routed through these systems but are not executed.) There must be huge demand for these dark pools with that volume pulsing through them…but is this really true?
One of the major (if not the primary) reasons a dark pool is “dark” or non-displayed is so that users may submit large orders and cross large blocks of stock without impacting the market or causing information leakage. Makes sense, right?
Well, recently, Justin Schack at Rosenblatt Securities printed an excellent article called “Let There be Light”. He did a review of all the top dark pools. Of the 50+ dark pools, only three actually trade blocks of stock. See for yourself. They are Liquidnet (52,000 avg. trade size in May), Pipeline (~46,000 avg. trade size in May) and ITG Posit (~5,000 avg trade size in May). Every other dark pool had average trade sizes at or below 600 shares! Do we really need a dark pool to trade 600 shares? Hmm….
Which really begs the questions, are dark pools other than the three mentioned above really adding any value at all (simply rearranging liquidity for commercial purposes); or worse yet, are they doing harm by making it harder and more expensive for investors to find true natural block liquidity? And what can we learn from the three dark pools mentioned above that enables them to realize the promise of a non-displayed environment?
The hint (which I will explain in another post) is that a dark pool needs more than opacity to put up blocks or even good trades and that something “more” may force us to rethink the importance of milli-second, high-frequency trading. Can a slower, more methodical electronic marketplace that aggregates non-displayed liquidity and/or permits qualified negotiation be a more effective solution?
Maybe we need to stop focusing on trading volume statistics and start thinking about market functionality and quality. And what are the real mechanics behind best execution in a non-displayed environment?
Well, my eyes are tired, I gotta put another round of drops in my eye and my vision is blurry but I am kinda feeling that I am on the path to seeing things more clearly. Hopefully we all are….
