Thomson Reuters announced earlier this month it had introduced RBOR to its REDI execution management system, enabling traders to set rules that determine which trades should be handled automatically rather than through manual intervention.
The most common use case involves dealing desks that process a large number of orders per day, many of which represent a small percentage of average daily volume, explains Michael Rude, the firm’s head of ETI trading.
“The dealing desk cannot add material value to these small orders and is best served by automatically routing these small orders to broker algos and staging the hard orders for the desk to manage manually,” he says.
“We have seen cases where dealing desks can take on additional mandates – such as a new asset class – due to the efficiencies that rules-based order routing brings.”
James Cusack, global head of sales at Caplin Systems, agrees that most banks will use RBOR for soft flow while the more challenging flow is still managed in some sort of book.
Brad Bailey, Celent
There are several reasons why traders would want to set order routing rules, notes Brad Bailey, research director with Celent’s capital markets division.
“Firstly, there is the desire to capture all trades – principal and agency – as well as channels of communication (electronic trading system, chat, voice) in front- and middle-office systems to ensure trade details are correct and risk is reflected,” he says.
“Secondly, there is flexibility in the workflow and channel in which orders are generated and communicated, and thirdly, it is easier working complicated cross-product, cross-market or cross-asset class trades.”
Automatic order routing can be especially effective for large volumes of smaller tickets in more liquid currency pairs, allowing resources to be focused on the more difficult tickets. Such tickets might be larger sizes, less liquid pairs or involve more complex trading objectives, allowing the skill and experience of the trader to be deployed.
For many types of trades, a rules-based approach demonstrates a clear plan for achieving best execution. It can be a valuable component of a firm’s best execution policy, and in a world of increasing pressure on costs and efficiency, RBOR should at least be considered, suggests Pete Eggleston, co-founder and director of BestX.
However, Integral chief revenue officer Vikas Srivastava observes that RBOR is only a part of what it takes to achieve best execution.
“Additional requirements include access to aggregated liquidity from all the relevant sources, a constantly evolving cloud-based execution system that delivers a comprehensive set of algos and optimal netting,” he says.
“Finally, you need a programme that is founded on deep insights from big data to tie it all together.”
Trading through RBOR might allow traders to focus on more demanding trades, and that focus could result in more efficient trading for those larger trades, but John Halligan, president of Global Trading Analytics, cautions that there are many variables involved in executing a trade efficiently.
“Best execution still depends on how effectively the trader uses the tools at their disposal,” he adds.
Different rules might be required during periods of low liquidity and/or high volatility or in trending markets, and identifying such regimes in a timely fashion is not straightforward.
It is also important to take into account the initial execution objective for the trade, says BestX’s Eggleston.
“For example, if a CTA [commodity trading adviser] is executing a buy order with the objective of minimizing slippage to arrival price, different rules may be required to an order from a passive index manager who is simply rebalancing a large portfolio shift,” he says.
An obvious technological challenge is ensuring that venue or counterparty changes are captured.
Celent’s Bailey notes that fixed rules require flexibility in times of market stress, or anything that changes the typical day-to-day flows, although he is confident that rules-based routing will evolve as advances in analytics – leveraging predictive machine learning – create a more flexible approach to routing logic.
Smart order routing should be combined with pre-trade transaction cost analysis (TCA) to determine the most appropriate order routing, suggests Barracuda FX CEO Kieran Fitzpatrick.
“When combined with independent post-trade TCA measurement, we can create a full feedback loop to achieve real best execution,” he concludes. “We see smart order routing becoming increasingly electronic and algo-driven, moving beyond the individual bank, but augmenting human behaviour rather than replacing traders.
“Effective routing rules reduce noise and distraction, leaving traders with more time to focus on exception management and higher-value requests.”