Waters Webcast: Aiming for Best Execution

Early last month, Waters hosted a webcast on the topic of best execution, sponsored by Bloomberg and Sybase, where a number of the variables pertaining to best execution—and by association algorithmic and high-frequency trading—in the US and European marketplaces were discussed. Panelists included James Rubinstein, director of algorithmic trading at UBS; Stephen Temes, managing partner and founder of Lincoln Capital; Mike Googe, TCA product specialist at Bloomberg EMS; and Stuart Grant, financial services business development at Sybase. The event was moderated by Victor Anderson.

Nebulous
As liquidity fragments across an ever-more diversified range of exchanges and dark pools, it has become imperative for firms to provide proof to their respective regulators, and, increasingly, their investors, that they are indeed meeting their best execution mandates. Yet best execution is a nebulous concept, with different sets of requirements in Europe and the US. In the US context it tends to be price-oriented, whereas in Europe it entails a wider range of variables.

In terms of the variables firms need to consider when providing clients with a service that qualifies as meeting best execution criteria, and the role technology plays in supporting such variables, Googe of Bloomberg EMS says the recent liquidity fragmentation phenomenon has created a number of data capture challenges.

“Establishing the parameters of each individual algorithm across all the different providers is a challenge,” he says. “And capturing the semantic of how those variables play out throughout the life of an order, and therefore making a relevant and accurate measurement, is also difficult. Fragmentation has to be taken into account as to how certain venues behave, but underlying that and tying back into the technology question is the underlying data set.”

The fragmentation of liquidity has introduced a number of data management challenges to market participants in terms of capturing the most appropriate prints so that they can make relevant comparisons between those various liquidity sources. “Do you want to see over-the-counter and do you want to see natural crosses in the calculation of your volume-weighted average price (VWAP) if you’re using that as a base-line calculation? Technology is working hard to capture that kind of information, but ultimately it’s down to the quality of data,” Googe says. “And challenges exist—specifically in Europe, but also across the globe—in terms of getting that accurately.”

Sybase’s Grant says content is paramount, with firms’ abilities to adopt a forward-looking perspective on the data needed, and devising optimal ways of gaining access to that data, underpinning their core considerations. “There are two issues,” Grant says. “First, how do you feed your current production systems? And second, how do you collect the information necessary to develop new approaches in the future that give you a competitive edge? We’ve found the average effective age of an algorithm to be around three weeks before it needs to be further developed, so companies need to start looking at other areas for opportunities, either for their customers or their trading partners.”

Challenges
In terms of the biggest challenges facing buy-side and sell-side organizations in driving their best execution strategies, Grant agrees with Googe that gaining access to the necessary information is the primary obstacle. “We saw after Mifid was initially implemented that organizations’ best-execution policies were limited by the technology they had in place and the venues they were able to provide connectivity to,” Grant says. “Consumer patterns have started to change a lot over the last 12 months—they’re either becoming more fickle about the product environment and the time span in which they’re able to execute, or they’re looking for opportunities that are less risky than they were in the past. So the biggest challenge is to be able to enact policies, that not only make compliance officers and regulators feel nice and warm and fuzzy about what’s going on within the company, but that also provide clients with a competitive advantage in terms of execution.”

The regulatory environment has led to the fragmentation of liquidity, Googe says, which has created complexity. “The answer to that has been for sell-side providers to get smarter about the kind of solutions they dole out. So the challenge is around keeping pace with that evolution, which seems to be on an upward curve,” he says.

MTFs and Dark Pools
Panelists were asked about the role that multilateral trading facilities (MTFs) and dark pools played in providing brokers with additional liquidity pools when looking to move large block orders, and whether misconceptions existed in the marketplace, especially from the perspective of intuitional investors, when it comes to using non-lit liquidity. Rubinstein of UBS says dark pools provide deep sources of liquidity, although caution is still needed when accessing such liquidity, especially as dark pools no longer provide the kind of no-impact trading they once did. “Misconceptions around dark pools can slide in either direction,” Rubinstein says. “A certain segment believes dark pools are a no-impact way of trading, enabling execution of a large block without impacting the market. That may have been how dark pools originally worked 12 to 18 months ago, but as they become ever more integrated into the workflow of algos, smart routers, and high-frequency strategies, you need to protect your order and information in the dark pool no differently than you would in a lit venue, and that’s something that we are always building our algorithms to work with.”

As far as the emotive issue of high-frequency trading strategies, and the extent to which such practices impact buy-side order flow—specifically that emanating from long-only managers—Rubinstein says that implemented correctly and intelligently, there is no reason they should be afraid of executing against the more urgent liquidity. “There is a time, a place and a price for any fill—it’s really up to brokers and the algorithms to spend time analyzing the liquidity we’re seeing in all the different dark pools and modeling the correct way to execute against it, depending on the algo, and therefore, the client’s objective at any point in time during the life of the order,” he says.

Steven Temes founder and managing partner at Lincoln Capital, says anonymity is one of the core reasons dark pools have come to the fore, even above price and execution, especially for larger institutions. “Particularly if they are an influential player in a stock, anonymity is the utmost consideration,” Temes says. “We observe that pretty frequently where we see a lot of volume trading in a stock and go to all the different brokers asking them what’s going on, but no one has any idea—the volume is there, players are there, and no one has a clue who it is.”

Satisfying the Regulators
The final question discussed during the 90-minute webcast, centered around the measurements firms use to satisfy regulators that they’re assessing and hopefully achieving best execution criteria, according to their individual policies. “The key aspect is being able to provide the policy that was in place at the time of a particular execution, and the ability to capture the market at the same point in time for that particular trade, or for that particular customer,” says Grant. “If a customer complains two years after the fact, how many organizations actually have the data available and accessible online to be able to recover the information within a timeframe that’s acceptable to the regulators?”

Googe says the answer to that question depends on the regulator and constituency you’re dealing with. “Certainly we have some very simplistic requirements we have to face up to, which are determined to show outliers where people have stepped outside the spread, which can happen to cross blocks for example,” he says.

“If you are a liquidity seeker rather than a price seeker, you need to identify those quickly and then tie that against the type of order you were trying to participate with at the time,” he say. “But we are seeing a lot more granularity now at the individual fill level and how those patterns have played up against the market data—it’s the kind of stuff you’d expect when prices have stepped out of line with the underlying market data. When you see in internal compliance the kind of activity that would be regarded as a conflict of interest—for example, where the same name has been bought and sold for the same account—those are the kinds of areas we are asked to look at.”