Increased choice is a blessing or a curse, depending on your quality of information. Without the ability to analyse the merits of rival products, the consumer is blind and bewildered by the options placed in front of him. Similarly, as buy-side traders begin to place a higher proportion of their orders in dark pools, they are also increasing their due diligence to deploy effectively the breadth of channels available to them.
“Dark pools are not for every one and certainly not for every order. Before deciding whether an order is suitable for a particular dark pool, risk and cost factors should be assessed by the trader who must do adequate pre-trade preparation,” says Stephane Loiseau, managing director, deputy global head of execution services, global equities and derivatives solutions, Societe Generale.
Traders should take full account of the objective of both the investment and the implementation strategy. In the first instance, the trader should consider whether he is a momentum player on a particular trade or is taking a more passive approach. Other parameters might include whether the stock is a short- or long-term investment, or whether there will be follow-on orders.
For example, a value investor with a buy order for a high percentage of average daily volume (ADV) in a small-cap stock with thin liquidity is likely to find a dark pool a more effective alternative to a lit market. For such an order, the displayed venue may prove costly because it is unlikely that the required volume of stock can be accessed quickly. Also, the risk of signalling information is less likely in a dark pool, not least because small cap stocks are not a high priority for so-called toxic flows, i.e. high frequency strategies operated by hedge funds. Overall, for this particular stock, there’s a good chance of getting a high proportion of the order done in a dark pool in a discreet and cost-effective manner.
But a buy order from a momentum player for a large-cap stock, with high volatility and sensitivity to news flow would be a bad candidate for a dark pool. First, there is an opportunity cost attached to the risk that the order may not be executed immediately, while sentiment on the stock in question changes rapidly in the displayed markets. Second, the order may be sending out information unwillingly to high-frequency strategies that do detect orders that rest in dark pools over a period of time. “You are sending out information, but you cannot react to that signalling risk because your order is resting in the dark pool,” says Loiseau. “If by contrast you’re in a lit venue and you’re calculating your estimated impact properly, you may be able to get that same order done in a shorter amount of time and with less signalling, because you’re in the driving seat deciding when you want to trade.”
Moreover, not all dark pools are created equal in terms of typical order size, variety of order flow, external links, order types or protection rules.
The trader should be aware of the way each individual dark pool operates; he must have a good knowledge of the dark pool microstructure. The trader must also protect the order against the well-known undesired effects of operating in a dark pool. “Some protection is offered by the rules of the dark pools,” says Loiseau. “For example, one pool is looking to establish a series of random crossing times throughout the day. This could offer strong protection for the end user. A random cross makes it difficult for a participant to interfere with the flow in the dark pool, because he can’t know when the cross will take place. This creates a neutral environment that is not focused on price formation, the traditional role of the lit market.”
But protection should also be part of the implementation strategy itself. An obvious level of protection comes in the use of a minimum order size, which removes the risk of pinging and gaming. An element of optimisation required: set the minimum order size to high and you reduce your chances of finding a cross; too low and you end up sending more information to the market and providing more opportunity for other types of flow in the pool.
Another important factor that can be often overlooked is the amount of time a trader is prepared to leave an order resting in a dark pool. “There must be a balance between waiting long enough to have a good chance of crossing but not so long that you leak information and risk increasing the visibility of the order through being pinged,” says Loiseau. “You can only find an optimal time of rest by comparing different dark pools and measuring your probability of crossing using standard parameters, such as trading style, stock type, order percentage of ADV, etc.”
Finally, it’s important to remember that dark pools are not regulated environments. They have their own rules and they are able to supply different levels of information to different participants. This is why the increased interaction between dark and lit liquidity is such a hot topic for regulators. Having invested heavily in trading systems and tools in recent years, buy-side traders may need to boost their pre-trade analytics capabilities to ensure they are not left blind-tasting in dark pools.
Please click on the link to register your vote on The Trade’s poll on dark pools