Zero Beta

Today it was announced that three major banks would link their dark pools of liquidity.

From FT Alphaville,

Goldman Sachs, Morgan Stanley and UBS are to link their private stock trading operations to improve liquidity and better compete with the increasing number of alternative exchanges. The move, to be announced Tuesday, will give clients of each bank access to the others’ so-called dark liquidity pools – the private interbank or intra-bank platforms widely used to trade stocks away from exchanges.
The pools are used by clients such as hedge funds to buy and sell large blocks of shares in complete anonymity and without the danger of moving the public price of a stock on an exchange. The development of dark pools is considered a potential threat to established exchange.

Now I must admit that I am relatively ignorant on the finer points of how these pools work, so any thoughts that I may have should be taken in that context. I have heard that they make up nearly 10% - 20% of trading volume, which is none too shabby.

I understand investment bank’s motives and can not fault them for providing this to their clients. First their is a strategic motive. They used to be paid for their access to public markets, but competition from discount brokers have killed this business. Then they were paid on their research and advice, but the internet and prior “conflicts of interest” have really taken a chunk out of those profits. Up until recently they have always provided capital - but the recent credit crunch and subsequent attempt to trim their balance sheets is hurting their growth in this area. In the meantime, exchanges have flourished as globalization has opened up markets and the growth of hedge funds and technological and financial “innovation” has created more trading and more products. In addition these dark pools are a way for them to get into the business by offering a private alternative. Second, it allows them to offer a “value-added” to their institutional clients. By giving them an anonymous, private form of liquidity, they provide something that many other can not.

While many think there is something sinister about these “dark pools” (with a name like that no wonder), in reality they are no different than the Auction Rate Securities - while they turned out to have problems in the event of a crisis, and I contend still were a form of bid-rigging, they were just a way for firms to please all their clients. Their failure was never planned, obviously, but the credit crunch came along and it was the only option.

The question that has to be asked is more of an ethical or philosophical one. Should these investors HAVE to move the market when the buy or sell a stock? There is an obvious transparency issue. Participants in this market are privy to information that others are not, Time and Sales, or just price and volume. This then implies that participants in the public market are unaware that there are more orders out there than what he or she sees on an order book. Those who have access to both markets have an edge on those who do not, and combined with an average system or strategy should be able to extract profits from those in the public markets. Large volume, in one direction or another is what moves stocks, or any financial instrument.  It is how the market discounts effectively.

“Having” to move the market when trading is the subsequent downside of someone who has the size, influence, or ability to influence the market.  Not having to move the market, enables someone of size to manipulate the physics of the market to their advantage.  This is the main point I wished to make.  What follows is an explanation that draws upon ideas in physics and chaos - topics that I am not an expert in.  I find it interesting, and think many of you will, but it is by no means a scientific conclusion, support for my main arguement, but rather an investigative hypothesis that sprung from this all.

Benoit Mandelbrot, in his works on chaos, fractals and markets, often talks about the idea of “trading-time”. For an good explanation check out this post on The Econophysics Blog. Basically, the idea is that in markets, time is flexible. During periods of great volatility and volume, time moves much faster, and during periods of relative low volatility and volume time moves much slower. Think of it in terms of your own life. Time seems to go much more slowly when you are at work staring at the clock, than when you are actively doing something. If you believe, as I and many others, in this concept of trading-time, then those with access to both exchanges are able to trade with the benefit of an additional dimension, scale, and are able to distort time and price.

To comprehend such a bold assertion, let’s look at the nature of the two volumes. The makeup of the volume in the dark pools is large orders that the investor wishes not to cause a market impact. These large orders, although they make up a small proportion of the overall volume, make up a large percentage of price movement. To be more exact, this volume, which is made of large orders put in by institutional investors who are viewed by the market as highly sophisticated, skilled, activist, determined to take a company private, and/or in possession of some inside information explain most of the variation in stock prices. Since this volume is made up large share purchases or large share sales, not a large amount of smaller ones, to each individual stock they are more significant and add more to the trend than they do to the noise.

This volume is what I call determined volume. The investors execute the trade with a certain goal or strategy that is known by the market and/or or will fail if the market should move against them. In the time frame of market-time, these investors can see ahead in market-time to some event that will affect price, but since time and price are interrelated, and the trade will alter price, they need a way to alter time and price. When executed on the public exchange, they move the market ahead in price, therefore in time. When done on a private exchange, they do not move the market ahead in price and or time. In essence, they are able to take advantage of price without distorting market-time, while others can not - virtually trading on another dimension than others in the system.  They are able to take advantage of scale in order to distort price.

This has important ramifications. If someone has access to both dark pools and public exchanges, there is no incentive for someone with any sort of quantifiable edge, or information to convey any information in the public exchange. There is, however incentive for them to convey false information. If you know a buyout is going to happen, or plan to buy out a company, you could purchase shares in the dark pool, dump them on a public exchange, and then buy them back in the dark pool at a discount. While there are supposedly moves to eliminate this “gaming” of the system, it is detectable only through algorithms that can easily be “gamed” as well. This will create then, a public exchange, where the motive of many trades is not for profit but to speed up or slow down the clock. In essence it will undermine the public markets all together, creating a reality that most could not comprehend. Market participants who access the public exchanges only will be trading with people with access to another exchange where there is greater volume.  These dark pools’ scale gives them the ability to alter time, without consequence to price exists. Even if used honestly, not “gaming” the system, every trade that occurs in the system will be one that does not effectively discount information, since any trades that would ordinarily discount the market are done at a discount.  Those with information will have an even greater edge over those who do not.  Information would discount at a much more accelerated pace in the public exchange, as the lack of these trades, affect the way the same amount of volume will affect time and price.  The price to access such an exchange would be substantial.

The two different exchanges causes two different trading-time continuum based on this idea of scaling.  In dark pools, where the average trade is 10 million shares, and there are participants seeking out various strategies, a trade of that magnitude becomes common place, and positions that would ordinarily alert the market to “something” means very little.  Only the persistence of such trades, or an abnormally large trade would seem out of place.  Therefore an observer of such a pool could gain very little relevant information from trades that are very material, and, if they did notice something abnormal would have very little opportunity to go to the public exchange to take advantage of such a thing since by the time abnormality is noticed, it would be too “late”.  All liquidity in the dark pool would be gone.  They could try the public exchanges, but scaling works both ways.   If all large meaningful trades were done in the dark pools, where they appear normal, the public exchanges would be a place where less meaningful smaller trades occur.  A trade that once seemed large, would now appear tremendous.  It would indicate to the market, just what it actually means - liquidity in the dark pool is dry - and the market would quickly move to reflect that.  Discounting would be more violent as well as accurate.  Trades could be “worked” over time, but it will take more time to effectively work them, as the market is more sensitive to volume, thus smaller blocks would have to be worked at the expense of time or price.  This will all have the effect of quick, large directional moves in the markets as volume from the pools spills over into the public exchanges, instead of dripping in.  What will result will be sustained periods of calm and little direction, followed by giant directional moves.

This phenomena plays right into many investors cognitive biases, as they tend to jump into high flyers and sell what isn’t moving. This is similar to what happened in the 1990’s tech boom when investment banks would underwrite crappy tech companies, only let a portion float, then as the stocks popped, float the rest of the shares causing a tremendous amount of volume to be absorbed in a short period of time, and stocks got clobbered.  In the case of dark pools, loads of volume can be gathered with very little impact, and once the market in the dark pool is alerted to this will become illiquid, and those who wish to buy more would have to move to the public exchanges.  As the public exchanges get woken up, and start catching up in market-time, those who bought the shares in the dark pool would be have a public liquidity pool eager to buy such a stock - its almost a self fulfilling prophecy.  This would make sense since those in the dark pool purchase stock in a higher volume exchange, where price and time move differently, and mean different things.  When demand spills over, for whatever reason, the force it exerts on price could cause mind-time to catch up to market-time,  causing more volume and force.  Eventually, there will be a period where there will be ample liquidity and enough momentum, aka volume, in the public exchange to conserve the momentum of a sale of the volume of shares obtained in the dark pool, and the advantage of the dark pool of liquidity will be reaped.  While not an expert on physics, I find many similarities to this phenomena and Einsteins theory of Relativity.

Dark pools allow two sets of investors to be subject to two different sets of rules in the market.  One set could cause a force that exhibits an equal and opposite force, but at a different time and price, when it works to his advantage rather than disadvantage.  It is lack of transparency in the dark pool but also the creation of two separate exchanges with different characteristics that enable this to occur.  While I doubt it is intentional, it still favors those with greater information and capital over those without.