iquidity in the U.S. equities market is dispersed among 50 trading venues, yet observers maintain that most of the action occurs in only a handful of venues. Finding liquidity and obtaining best execution without leaking information involves advanced mathematical modeling and network optimization. Asset managers are finding they are only as good as their routing algorithms. The displayed markets and “dark pools”* offer varying degrees of transparency, so the networked trading landscape has become extremely complex. Asset managers are challenged to come up with new, innovative ways to reaggregate the marketplace and to reestablish some rationality for the routing and trade process. Five years ago, the routing table was the primary means by which brokers/dealers decided how to distribute orders. When an order was received, if multiple quotes were available for a particular price level, the broker/dealer would consult the routing table to determine the order in which the various venues would be accessed. The next step would be to use a series of routes to sweep the marketplace and ask for liquidity at the various venues. This approach was effective at a time when liquidity was concentrated in a finite number of venues. But routing tables have become increasingly inefficient, especially when it comes to accessing liquidity in dark pools. Tracking liquidity and responding to it is a dynamic process. While the tables might change from day to day and stock by stock, they tend to be static on a given day for a given stock. For asset managers, determining when and whether to interact with a liquidity provider has become a significant challenge. “Most clients have spent a lot of time and effort determining what their success rates are,” says Laurie Berke, principal at TABB Group, a New York-based research organization. “Liquidity is somewhat fluid, so it does shift around a bit between the various venues, both lit and dark.” The underlying investment strategy drives the actual trading strategy. Asset managers who trade illiquid microcap securities have to be especially careful not to leak information and let their orders have an impact on the market. Their success rate typically varies between venues, and their experience will be different from that of a large-cap growth manager. Some asset managers use aggregators that tap into multiple venues and search for liquidity in real time. Others send orders directly to specific venues. Block orders can be sent to a crossing network—such as Liquidnet, Posit, or Pipeline—to be executed directly with another asset manager. Those who need a fast fill on highly liquid orders may not be willing to sit in a crossing network, so they might use an algorithm and a smart-order router to tap into the various pools. “Based upon the current success rate, as well as historical success rate for that type of security, the algorithm and the smart-order router will be a faster, more efficient way to find out where that liquidity is right now,” says Berke. EdgeTrade, a direct market access agency-only algorithmic specialist owned by Knight Capital Group, offers P R O F E S S IO N A L P R A C T IC E three smart-order execution tools: FAN, Covert, and Sumo. According to Joseph Wald, managing director of Knight Capital Group, buy-side clients consider smart-order routing tools essential in the current volatile markets. Moreover, they want consultants to help them achieve best execution. To this end, Knight works closely with its clients to understand their portfolios before developing custom algorithms for them. Clients are also asking for algorithms within algorithms. For instance, Knight offers an algorithm that tracks a percentage of volume. Within that algorithm is another algorithm that limits a trader’s orders to no more than, say, 10 percent of the volume. “The reason that trader is doing that is not because he wants to be 10 percent of the volume,” Wald explains. “It’s because he believes that being any more than 10 percent of the volume will adversely affect his execution performance and create a lot more market impact or maybe tip the marketplace about the order he’s trying to work.”
Read full abstract
Jul 22, 2011
Frederick H Deb Harris + 2
Open Access