As the equity markets grow increasingly more complex -- characterized by fragmented trading venues, high-speed trading and accelerating volatility -- the buy-side is adopting more sophisticated transaction cost analysis tools to match. In fact, more and more buy-side firms are tapping new TCA capabilities to monitor and fine-tune their strategies intraday.
While TCA has long been a necessary buy-side compliance tool for proving best execution, it now is gaining momentum as an alpha-preservation tool.. Buy-side traders are seeking a new era in transaction cost analytics that can monitor the performance of algorithms in real time; accurately predict the cost of accessing liquidity in a more frenetic, high-frequency market; and even select which algos will maximize returns.
According to a May 2010 Celent report on TCA, the financial crisis and the explosion of high-frequency trading have changed the rules of the game for the buy side such that adapting execution strategies in real time has become necessary to maintain returns. While explicit trading costs (i.e., commissions) have declined and spreads have narrowed, Celent reports, implicit execution costs -- that is, market impact, timing, delay, liquidity risk and price volatility -- are increasing due to the presence of high-frequency trading firms, which leverage the fastest market data feeds, pattern recognition software and information arbitrage. To make more informed execution strategy decisions -- such as which brokers to use, which algorithms to select and which dark pools to turn off -- buy-side firms must use TCA, the research and consulting firm asserts.
The Future of TCA
"If you're not running [intraday] calculations, you're coming to a gun fight poorly armed," warns Fred Federspiel, CEO of Pipeline Trading Systems. To help traders react to real-time market conditions, Pipeline developed what it calls an Algorithmic Switching Engine that predicts the performance of algorithms under real-time conditions based on an information feedback loop.
"It takes a trader's instructions about how hard the trader would like to drive an order into the market and translates that minute by minute into a choice of an algorithm style, price limit for an algo, the control parameters on that algo and the size of the child orders sent to that algorithm, and then we send that out and let the algorithms execute it," Federspiel explains. Currently, Pipeline's switching engine reaches out to 132 algorithms and is adding about 10 more each month, Pipeline says.
According to the Celent report, predictive analytics that weigh adverse selection costs -- such as Pipeline's Algorithmic Switching Engine -- are the future of TCA. From January 2009 to May 2009, AllianceBernstein conducted a joint study with Pipeline Trading to define "adverse selection" and measure the effectiveness of predictive switching on adverse selection costs. AllianceBernstein used a dark aggregator from a different sell-side provider while Pipeline ran its Algorithmic Switching Engine.
"We suggested a way to quantify adverse selection, so we took a relatively straightforward dark aggregator strategy and modified it to minimize the effects of adverse selection," according to Dmitry Rakhlin, global head of quantitative trading at AllianceBernstein, a U.S.-based investment management firm majority-owned by Paris-based AXA. With less development effort than required to implement the switching engine, AllianceBernstein improved the performance of the algorithm, explains Rakhlin. "The performance has improved significantly, but not to the degree of the switching engine, which incorporates a richer short-term predictive model," he relates.
The study, according to a 2009 Pipeline research paper, showed that using Pipeline's predictive switching strategy -- as compared to a common dark aggregator strategy -- saved as much as 70 percent of the costs attributed to adverse selection.
One buy side trader who requested anonymity suggests traders can even set parameters for predictive algo switching and forget it. "Once traders are comfortable that the switching engine is performing well, they can outsource minute-by-minute decisions to the switching algorithm," he claims. "You can still impose your controls, put your limit price and you can decide how much [of an order] to expose to the switching engine -- then it shows superior performance."
Tim Sargent, president and CEO of Quantitative Services Group, a Naperville, Ill.-based independent global provider of TCA, says, "Institutions have realized that traditional transaction cost analysis, or TCA, doesn't give them information that helps them determine intraday trading and the impact of high-frequency activities." According to Sargent -- who argues that with the prevalence of high-frequency trading, buy-side firms need to separate the cost of their own algorithmic trading from the trading going on around them -- "One of the biggest problems is that TCA techniques haven't evolved with the market, so people are using antiquated ways of evaluating execution quality."