The Evolution Of Asian Algorithmic Trading


With Murat Atamer
How has increased algo trading affected market microstructure across Asia?
Market structure, regulatory environment, and advanced technology are the key factors promoting the use of algorithms. This is evident from the fact that a good trader uses drastically different algorithms in South East Asia from those she would prefer in liquid developed markets in our region. Spreads are the driving force as once they get smaller, displayed liquidity decreases and manual trading becomes difficult, resulting in an increased use of algorithms.
Over the past few years, we have seen a considerable increase in the portion of overall trading going through algorithms. This is evident in the data as volume curves across securities within a market look increasingly similar to each other. The fact that VWAP and volume participation algorithms have been the “go to” algorithms for most buy-side traders explains these patterns.
Recently we see a heightened interest from the buy-side on opportunistic algorithms that go beyond their predictable, yet very mechanical, predecessors. Traders want algorithms that “act” like a human trader by focusing on real-time market dynamics. There is also an increased focus on performance and a move away benchmarks that encourage high participation rates.
What comes next for algo strategy and development, and who will drive the change – regulators restricting innovation, buy-side pushing for more control?
These are exciting times. The SFC’s regulatory drive forced traders to get a deeper understanding of the algorithms they are using. This has opened up a lot of debate between the algo users and the designers. We have seen a decline in the number of algorithms used by a typical buy-side trader. All of a sudden safety measures available to brokers to protect buyside orders have taken the front seat. Traders were more interested in fat finger protections and on how broker circuit breakers work than getting another custom algorithm; and therefore, for the first time we have seen cancellations on custom algo requests from the buy-side.
The regulatory push for safety has set the stage for less algorithm providers as well as less algorithms. This is likely to mark the beginning of a period for further specialisation by the traders as well as by brokers where they will focus extensively on bettering their existing algorithms. A much greater level of transparency between the buy-side and the sell-side is crucial for achieving this systematically. Particularly, the traders have no visibility on the child slices sent to the market by the algorithms that they are using, and, therefore, have been pushing the sell-side on fix tags that provide liquidity indicators, i.e. Tag851, which would provide this information real-time. Electronic trading platforms that provide feedback to designers as well as the users will naturally evolve to be superior to mechanical platforms that lack this capability. The feedback does not necessarily need to be real-time as long as it is available for in-depth analytics. In this regard, TCA can be a crucial tool to bridge this gap and provide the much needed transparency to the buy-side. Historically, TCA has offered little insight in terms of “selecting the right algorithm.” With this in mind, we have been actively looking at our platinum client orders and have been closely working with them on their algorithm choices based on stock, market, as well as order characteristics to help them achieve better execution.