Dr. Paul Lynch has been at the forefront of electronic trading for over 20 years, having started his career as a Liffe trader, ran algorithmic trading at Citigroup and founded Itarle in 2005.
Yet the industry has changed beyond recognition over the past two decades, as post-2008 financial crisis regulation has piled the pressure on investment banks in particular.
Stricter capital requirements and prescriptive trading rules such as Europe’s Mifid II have forced banks to reconsider the instruments they trade and how they trade them.
Mifid II, which took effect in early 2018, required all firms that trade European products to comply with its rules and, given virtually all banks and brokers above a certain size have an interest in Europe, Mifid II has in the past two years effectively become the global standard.
Dr Lynch, the chief executive officer of Itarle, said: “The wave of regulation has meant that banks and brokers are really focusing on the efficient allocation of capital to where their core strengths are.
“We are increasingly working with banks and brokers who want to partner with an expert in best execution algorithms that allows them to retain their own corporate identity through using a best execution algo service that has expertise in market microstructure,” he continued.
One of the most draconian of the many Mifid II requirements is that firms must show regulators they are getting the best deals for trading customers, an idea known as best execution that relies on transaction cost analysis (TCA).
Dr. Lynch said: “When it comes to TCA, should banks be marking their own homework? Mifid II has mandated TCA but does not require a third party to handle the analysis. Different banks have taken different approaches – some have made a point of the fact they have a third-party handling their TCA. We will see this debate evolve throughout 2020 and beyond.”
Itarle, as a supplier of customised trading algorithms, is well-placed to act as a third party, helping clients demonstrate their compliance with Mifid II’s best execution requirements.
The firm, which is based in Switzerland and has had a presence in Asia since 2007, sees Asia as a major growth opportunity over the coming years as Mifid II becomes more pervasive but Dr. Lynch admits it is a complex region to navigate.
The Itarle chief said: “Asia does not have a single exchange model; it is highly fragmented and different to the US or Europe, which are becoming increasingly homogenised.
“In Asia, the challenge is understanding the market microstructures. Different instruments have their own characteristics, which only become apparent if you listen to the order book. There is a lot of work tuning the algos to individual markets and Itarle has invested a lot of capital in this.”
Itarle of course is not in the business of building algos for every contract out there, rather it is looking for liquidity of at least 50 ticks a day which ensures enough data for the firm to “build a microstructure model with which to calibrate order placement techniques”.
The firm goes further than rivals by building customised algos for each maturity of a contract, capturing more of the spread on each instrument and delivering superior execution performance.
Looking ahead, Dr. Lynch sees complex options strategies, such as straddles and strangles, as an opportunity for his firm.
He is also looking to use Itarle’s algos to help banks and other firms to make the best use of the scarce collateral they have.
Dr. Lynch said: “As we are now some two years into Mifid II and the dust is starting to settle, we are looking at using algorithms to help banks reduce margin. Futures exchanges remain competitive and there is a strong and increasing appetite among both the buy-side and the sell-side to reduce margin.”
The Itarle chief executive said the firm is also looking at building a real-time market surveillance solution, which would be an improvement on the current crop of surveillance systems.
“If there’s a problem, we believe that people would rather know immediately. The first wave of surveillance was done out of necessity but, with advances in real-time machine learning logic, we certainly see these technologies playing a role in market surveillance.”
Dr. Lynch added: “Market surveillance involves looking at metrics such as order to trade ratios or size displayed versus fills. We need to be listening to the trader in real-time to see if his or her behaviour is different to what was happening an hour ago or a week ago. This requires breaking the individual complexities of the different trades and user types.”
Contact person: Edica Lin
Tel: +852-3782-0002 (O) / +852-6353-3199 (M)