Stéphane Leroy Chief Revenue Officer at QuantHouse
Stéphane Leroy Chief Revenue Officer at QuantHouse

How AI is helping FX mature as an asset class

Stéphane Leroy talks about how FX markets are currently within a period of intense, rapid transformation, meaning now is the time to take advantage and apply next-generation technologies such as AI and machine-learning

Trading foreign currencies is actually a relatively recent innovation. It has only been since the collapse of the Bretton-Woods agreement in the 1970s that rates of exchanges became a free-floating variable instead of a fixed rate regime. Since that time, FX has been traded almost exclusively in the OTC format as trading in the forex market is not organized via any centralized exchange like you have in equity or futures markets for example. All the different types of forex products that traders or investors trade in is always via market medium or a market maker. This is commonly a bank or a forex brokerage firm that helps facilitate your desired trade and helps you buy and sell specific quotes from customers and take orders from them. As a consequence, this asset class wasn’t ready for the Algo revolution.

Over the last couple of years, it has been very slowly begun to be viewed as a tradable asset class in its own right by Automatic trading buy side firms, as well as being used for real-time valuation and risk management, rather than to simply enable portfolio valuation.

Today, structural shift and rapid evolution in the market mean that FX is now following the footsteps of other more established asset classes such as equities and futures. Indeed, JPMorgan recently reported that FX algo transactions on the bank’s trading platform had risen 47 percent in 2017, while the number of users climbed 41 percent. We examine how automation and innovative technologies are helping push the popularity of FX as an asset class in its own right.

Growing demand for automated FX markets

Against the background of interest in equity markets drying up, as well as ongoing global diversification away from the almighty US dollar, the popularity of currency trading is being driven by a combination of trends including regulatory change, new technologies and new Fintech entrants intent on pushing through further market structure changes. 

In the past, price reference points in the FX market did not exist due to fragmentation and limited price transparency. Price formation was dictated by banks in what was historically a peer-to-peer discussion. Changes in market structure due to regulations such as MiFID II have increased overall demand for price transparency and capabilities to evidence best execution. In addition, cross-asset correlation, hedging and trading have set a clear demand from the buy-side for new price feeds through APIs across the FICC markets. Because of the rise of the robot and the existence of electronic pools of liquidity, pricing can now be monitored through fast automated systems in real-time. This opening up and monitoring of pricing effectively means that the enablement of AI, machine learning and big data can now be efficiently implemented for the benefit of the overall industry.

A market in evolution

Fast, automated processing is the key to leveraging trading opportunities, and automation of the currency markets with open prices is beginning to convince leading hedge funds that FX should be viewed as an asset class. This in turn should lead to an increase in more volumes being traded. In addition, new entrants using automated tools are acting as a catalyst to help the market embrace wider acceptance of FX trading. For example, the further and ongoing contribution on pricing of currency pairs helps to enable a constant price reference point for the industry. This has been evidenced by FastMatch’s 2017 launch of a global consolidated tape for the FX market. The tape includes transactions from a wide variety of market participants with 80% of the volume coming from trades not executed on the FastMatch ECN delivering, for the first time, a reference point for executed trades in the spot FX market.

Because of the rise of the robot and the existence of electronic pools of liquidity, pricing can now be monitored through fast automated systems in real-time.
Because of the rise of the robot and the existence of electronic pools of liquidity, pricing can now be monitored through fast automated systems in real-time.

Simply stated, the FICC markets are expected to mirror the operation of the more mature equity and future markets that we know them today. JPMorgan’s report points out that clients prefer FX algos because they can cut costs while allowing traders to better track and report their performance. It also noted that for investors who already have know-how in equities, the ramping up in currencies can therefore happen quite quickly.

It is also noteworthy to point out that if algo trading becomes the norm in the future, the FX market will be affected not just in terms of size, but also the way that humans are managing this market from “sentiment driven” to “data-based decisions”.

Take advantage of change

The common theme behind the various drivers across the FX markets is ultimately price transparency. There is a growing demand from the buy-side to be exposed to FX feeds, and brokers are keen to service this by ensuring their feeds are available in trading ecosystems.

Firms need to start aiming to be ahead of the curve of market evolution within the FX market, as well as acting as a catalyst for change on behalf of their respective customers. FX markets are currently within a period of intense, rapid transformation, meaning now is the time to take advantage and apply next-generation technologies such as AI and machine-learning to enable the inclusion of FX as an asset class in trading strategies.