Over the past decade, while public attention was focused on the rate of electronic trading adoption in the equities market, enormous market opportunities cropped up in other asset classes. The emergence of foreign exchange (FX) as a legitimate asset class has resulted in rapid adoption of electronic trading in the FX market, led by the tireless efforts of various inter-dealer, multi-dealer, and single-dealer platforms to create greater market transparency and support for full the life cycle of an FX trade.
In recent years, the FX market has witnessed the emergence of a new trend in electronic trading: algorithmic trading strategies designed to capture execution opportunities in an increasingly automated and fragmented marketplace. But as the market reality for the FX market continues to evolve, it is important not only to assess the potential for growth in adoption of FX algorithmic trading, but to identify possible hurdles.
Hurdle #1: Increased latency It is undeniable that electronic trading has increased substantially over the last few years in the FX market. With the increased participation of various proprietary trading desks (both bank-affiliated and independent) and actively trading hedge funds, speed of execution is increasingly becoming a major issue in the FX market, which by definition is global in nature.
This means that there is no centralized location for execution. A trader attempting to trade EUR/USD might be looking to capture market data from multiple locations scattered around the different time zones to ensure optimal trading execution. This is a major issue for actively trading firms. Depending on the distance of the various potential execution venues, a certain amount of latency exists (purely based on difference in distance as well as the existing IT infrastructure of the various execution venues), which can seriously hamper an overall trading strategy in which every millisecond counts. This latency issue is expected to get worse as electronic trading adoption increases, and the FX market continues to attract high-volume, low-frequency trading firms.
Hurdle #2: Resistance from FX Banks Beyond the latency issue, the ultimate future of algorithmic trading in FX is murky at best. On one hand, there appears to be a big push from leading broker/dealers (and not necessarily the traditional powerhouse FX players) to leverage the experience gained from the equities market and apply execution-centric algorithms to the FX market. On the other hand, vested interest among the major dealing FX banks makes them view execution-focused algorithmic trading with a healthy dose of suspicion.
Leading FX banks have developed sophisticated e-commerce engines so their prices can be streamed into various ECNs and other execution venues. Banks have been using these advanced e-commerce engines to not only internalize uncorrelated order flow, but also to accept flow from high-frequency proprietary trading shops (which would have been considered "toxic flow" not too long ago). Banks now have the ability to shade their prices more accurately, and a few of them even offer fractional pip pricing, which could benefit both the bank and the actively trading prop shops and hedge funds. As the influence of hedge funds has grown over the years, ECNs such as EBS (now part of ICAP) and Reuters, which previously only served the interbank market, have opened up their platform to direct hedge fund participation, thus blurring the lines between the interbank and client-to-bank markets.
As market fragmentation grows, fueled by ECN competition as well as individual bank platforms, market aggregation becomes more important. In fact, it has become one of the most important developments in recent years. Independent trading platform vendors have emerged to meet this growing market demand from clients. In addition, large dealing banks have increasingly turned to internalization to reduce costs and improve profit margins. More specifically, internalization enables banks to capture the spread internally and avoid paying the spread to an ECN, leaving virtually no footprint in the marketplace and eliminating ECN brokerage fees and associated interbank settlement fees, such as CLS costs.
While there are certain signs for changes in the marketplace to accommodate the growing adoption of algorithmic trading, Aite Group is not convinced that today's execution algorithms are being marketed to the right type of client segments. In today's fairly immature FX execution algorithm market, realistic users of broker provided execution algorithms might be corporations or certain types of asset management firms that can afford to wait for a fill and not be hampered by speed of execution.
In fact, Aite Group interviews with actively trading buy-side firms indicate that most of the focus to date has been on investment algorithms and the type of support they are looking for from their banks. This includes access to credit, liquidity, competitive pricing and speedy transactions. Most buy-side firms interviewed did not fully understand the value they would receive from broker provided FX algorithms, as they view the FX market as incredibly liquid, and few have faced major issues moving large size cost-effectively. Leading trading technology vendors in the FX market — including FlexTrade, Portware, Aegis Software, TradingScreen and Apama — have been leveraged by the various buy side firms for creating a more aggregated view into the various execution venues, including ECNs and single-bank portals.
While algorithms have certainly found a place within the FX market, it is not clear who will be leading the charge in terms of driving the mainstream adoption of FX algorithmic trading. To date, the clients and proprietary trading desks of dealing banks have taken the lead development of FX investment algorithms, but for their own consumption only. Despite the similarities that exist between the equities and FX markets, the sheer liquidity of the FX market appears to be vast enough that the prospect for client adoption of broker-provided plain vanilla VWAP/TWAP-type algorithms seems highly unlikely, especially involving the most liquid currency pairs. One area of interest in the long-run could be algorithms developed for less liquid and/or emerging market currency pairs. Despite these potential pitfalls, most banks and broker/dealers are diving head-first into the FX algorithmic marketplace. Unfortunately for most of these players, the concept of "build it and they will come," might not actually apply when it comes to the algorithmic trading in the FX market.