By Sylvain Thieullent, CEO of Horizon Trading Solutions.
With dozens of headlines circulating this year around the threat of emerging technologies automating swathes of banking roles, you could forgive colleagues on execution trading desks for feeling a tad jittery. The unrivalled speed and precision of trading algorithms have given rise to rumours of redundancy for years, and reports from the likes of Citi claiming more than half of banking jobs could soon be replaced only add to the sense of unease permeating the industry. But do execution traders really have anything to fear?
On the surface, execution trading – which traditionally involves buying and selling large quantities of stocks to complete a financial order – does seem a task that algos can handle more efficiently than humans. After all, machines don’t tire, they can analyse vast amounts of data in nanoseconds, and they don’t make emotional decisions. It is hardly surprising that they already handle the lion’s share of trading. According to Deutsche Bank, about 90% of equity-futures trades and 80% of cash-equity trades are executed by algorithms without any human input. Surely, one might reasonably assume, it won’t be long before machines snatch the last slither of trading action from their human counterparts.
But that’s an oversimplified view. Automation in trading is not necessarily about replacing human traders; it’s about evolving the role. In the late 2000s, when high frequency trading really took off, countless banking execs prophesised the end of bread-and-butter trading. More than a decade later, their vision hasn’t materialised. Why? Because as tech becomes more sophisticated, the execution trader’s skill set doesn’t disappear, it adapts.
In the 1990s, long before the SEC’s Regulation National Market System (Reg NMS) galvanized the transition to electronic trading, an execution trader’s job primarily involved physically pushing orders through on the floor, tracking prices, and executing trades in a fast-paced, often chaotic, environment. This required a deep understanding of market movements, timing, and, most importantly, human intuition.
However, the role of the execution trader in modern markets looks a good deal different. Execution management systems are more advanced, but the complexity of markets and the need for strategic decision-making is also greater. Today’s traders must shift from being order-placers to sophisticated managers of algos, copious amounts of market data, and, of course, risk.
Real-life traders now need to understand not only how the algos work, but how to optimise them in different market conditions. This requires knowledge of coding, data analytics, and an in-depth understanding of how different trading strategies — both human and algorithmic — interact with each other. In other words, traders need to be tech-savvy, rather than manual executors.
Additionally, while algos are exceptional at handling repetitive tasks, human oversight and judgment remains irreplaceable in many situations. When left to their own devices (if you can pardon the pun) high frequency trading algorithms can exacerbate plummeting equity prices by rapidly selling off positions. The 2010 flash crash is a prime example, while, more recently, leading analysts have warned such sharp selloffs could occur even faster.
Algos are also designed to operate within certain parameters. They can’t easily adapt to extraordinary market conditions or black swan events, and they are poor at interpreting market sentiment — something that can shift dramatically with a piece of breaking news or unexpected geopolitical event.
In these moments, a trader with strong market experience and an understanding of behavioural finance can intervene, reprogram the algos, or manually override systems to navigate through volatile periods.
Consider the airline industry, where automation in the form of autopilot technology has handled much of the routine control since the 1980s. Pilots remain very much essential in navigating complex situations and unexpected turbulence. This relationship is even evident in the most complex aerospace manoeuvres. SpaceX’s rockets now draw on autopilot AI to land projectiles in precise ocean locations once the astronaut’s role is over, permitting the spacecraft manufacturer to reuse its rockets. This is yet another clear example of how alternating phases of automated processes and critical decisions made by humans can lead to optimal outcomes.
Similarly, execution traders play a crucial role in guiding trading strategies through unpredictable markets, interpreting sentiment, and making real-time judgment calls beyond the reach of automation. Just as pilots and autopilot systems have worked in tandem for decades, human traders and algorithms complement each other, each excelling in distinct aspects of the trading process.
Look no further than the Covid-19 pandemic-induced market volatility of March 2020 for a prime example. One word that cropped up time and again as governments began locking down was unprecedented. If algorithms could feel fear, this word would surely stir it in them. While algorithms excel at trading in benign markets – and even tough situations for which data has been amassed – uncharted waters prove extremely perilous. This is where people prevail. Human traders can assess the situation more holistically and make judgments that go beyond pre-programmed rules. It is in these moments of uncertainty, where intuition and experience must stand in for data, that human traders will always provide value.
Unless novel market scenarios are behind us, the future of execution trading cannot be viewed as man versus machine. Man with machine is more fitting; co-bots, not robots. Execution traders and tech will work together in a symbiotic relationship. Algos will handle the mundane, high-speed, data-driven tasks, while human traders will focus on developing strategy, risk management, and the nuanced decision-making that machines can’t replicate. The role of an execution trader will increasingly involve collaboration with data scientists, software engineers, and quants. It will be about deploying tech to be more effective, not competing with it.
Automation will indeed transform execution trading. But it won’t erase the need for skilled professionals. Instead, the execution trader of tomorrow will look more like a tech-empowered strategist than a button-pushing trader of yesteryear.
Subscribe to our newsletter
