Conference Dailys TRADETech Daily 2019 - Wrap-Up Issue - Page 14

THETRADETECH DA I LY in-depth THE OFFICIAL NEWSPAPER OF TRADETECH 2019 THETRADETECH DAILY in-depth THE OFFICIAL NEWSPAPER OF TRADETECH 2019 Industry slams regulatory ploy to shift trading volumes to lit venues Patience is a virtue when it comes to machine learning say experts MARKET PARTICIPANTS AGREED THAT TRADING VOLUMES HAVE FAILED TO SHIFT TO LIT VENUES UNDER MIFID II, BUT THE INDUSTRY EXPERTS TALK DOWN IMMEDIATE RESULTS FROM MACHINE LEARNING AND WARN THAT THERE ARE NO SHORT- REGULATION HAS LED TO UNINTENDED POSITIVE DEVELOPMENTS IN THE FORM OF PERIODIC AUCTIONS. CUTS TO SUCCESS. S everal industry veterans have criticised efforts by policy makers in Europe to force more trading volumes from dark to lit venues under MiFID II. Senior market participants from European exchanges and the sell-side agreed during a panel discussion that the central aim of MiFID II to shift activity to on-exchange venues was misguided, although it has led to positive innovation within the industry. “I was surprised that the unintended conse- quences of MiFID II have actually been positive. Before MiFID II was implemented, we were all convinced that it would have a negative effect on markets, but instead we’ve seen positive innovation driven by commercial need,” said Richard Semark, CEO of UBS MTF. “The policy’s objective has been to move trading to lit markets, with no backing or basis for that. But the reality that we see as market participants is that lit venues are often not the best place trade, so other venues have come into the marketplace because we are still committed to getting the best outcome for our investors.” Similarly, Mark Hemsley, president of ex- change operator Cboe Europe, told delegates that investors are constantly seeking low impact execution, previously provided by broker crossing networks, which have been shut down under MiFID II. The closure of broker crossing networks has seen a rise in the use of alterna- tive trading venues such as periodic auctions “The policy’s objective has been to move trading to lit markets, with no backing or basis for that.” RICHARD SEMARK, UBS MTF and systematic internalisers (SIs). “If you speak to the buy-side and brokers, there’s always been a genuine need to execute in low impact ways through dark pools or broker crossing networks. Those venues have clashed with regulators forcing the political drive to move everything towards a lit environ- ment,” Hemsley said. “But the industry finds a way around that. The problem we saw moving from MiFID I to MiFID II was the incremental approaches to dark trading, and the artificial double volume caps have led to a surge in activity within the SI environment and periodic auctions, ulti- mately detracting from lit liquidity. Regulators need to take a more fundamental approach.” The regulatory aim to shift volumes to lit markets was a keen topic of discussion at TradeTech Europe this year, with Citigroup’s head of European market structure, Jame Baugh, stating on a panel session during the first day of the event that navigating the new trading landscape has been a huge effort for the industry. “A lot of innovation, time and resources have been spent on looking at ways of sourc- ing liquidity, connecting to SIs, looking at periodic auctions and navigating the liquidity that they can provide,” Baugh said. “The unfortunate outcome is that the channel shift has been fairly muted. When we look at the percentage of business that has migrated to lit venues, it has been relatively small.” On lingering concerns about the UK’s impend- ing exit from the European Union, some of the panelists agreed that the long-term impact of Brexit could be positive in terms of increasing competition. But others reiterated the need to find sufficient harmonisation with some form of equivalence to preserve the UK’s right outside of the EU to be more creative. “It’s about finding that balance because ulti- mately we want integrated capital markets. I agree that increased competition will be a positive, but we don’t want to jeopardise equivalence,” Rob Boardman, CEO of Virtu Ex- ecution Services for EMEA at Virtu Financial, said.  C apital markets firms that are looking to implement machine learning and artificial intelligence (AI) systems within their trading processes must be prepared to undertake a multi-year project that requires significant patience before seeing results. Industry experts taking part in a keynote inter- view outlined how their firms had approached machine learning projects and warned that those expecting immediate results from such endeavours would most likely be disappointed. Antish Manna, head of execution research at MAN GLG, part of MAN Group, said that the firm went live with a machine learning-based framework for order flow and broker allocation last year. “This framework effectively takes away the need for humans to set an arbitrary target for 14 THETRADETECH DAILY “There is a perception that you can hire people and have meaningful, AI- based outcomes…it doesn’t work that way.” SHARY MUDASSIR, RBC CAPITAL MARKETS ‘my first three brokers are going to get this amount of flow’ and continuously updating that target to having a machine that automatically does that”, Manna explained. “The beauty of it is that it becomes a very clean conversation with our brokers; they know how we are doing things and that they will get more flow, and this machinery also adapts to changing market conditions.” Manna explained that although the hype around machine learning has grown to a point where expectations are now becoming unre- alistic as to what the technology can achieve, starting with a relatively simple element such as broker allocation means the firm can build out the framework to take on more expansive and intuitive projects in future. Representing the sell-side was Shary Mu- dassir, co-head of global equities execution at RBC Capital Markets, who agreed that industry perceptions around machine learning were often false, particularly around how long such developments take to complete and the amount of time it can take to acquire the required expertise. “There is a perception that you can hire people and have meaningful, AI-based outcomes… it doesn’t work that way. Real success with AI requires very large teams,” he said. “At RBC, we’ve got in our AI research team over 100 AI scientists. Within the applied AI space we have over 300 data scientists on the bank side. On the equities execution team now, for the most recent product we will be rolling out at some point over the year, we have a team of over 60 people, and these 60 people are not all AI scientists – these are sales traders, traders, quants, execution consultants, technol- ogists…they have all come together over that period of time to deliver on one big outcome.” Addressing those unrealistic expectations, Manna said that the majority of time spent on machine learning projects is used to clean data before research and development can take place, and that those firms that are only now starting their journey with machine learning should be not expect to see results in the short- term. “The truth is, it is a fallacy and it takes a huge amount of time to build a framework where you can deliver things at scale that work,” he said. “On the machine learning and AI side of things, problems are best solved by teams of people, because you need the challenge, rigour and time to learn and fail, learn and try again; that process takes a lot of time.” Issue 2 15