With the hiring of Christopher Matsko, Head of FX Trading Services, FactSet made a conscious decision to pivot from its original FX client base of systematic trading Hedge Funds to focus more purely on the firm’s core client base of traditional asset managers and sovereign wealth funds. The refocused strategy, which leverages Matsko’s extensive prior experience in the real money e-FX trading business, has positioned FactSet at the forefront of trading solutions for institutional foreign exchange. The firm has won a number of industry awards over the past several years and begun to take aim at further expansion in Europe and Asia-Pacific.
We spoke to Christopher to discover more about how, under his direction, the firm is now expanding its technological footprint by opening liquidity channels and execution methods not previously accessible to the long-only trading sector. These new tools will lead the industry in front- and middle-office automation capabilities and revolutionize the way trading desks and middle office operations function today.
How is FactSet planning to change the face of FX Trading?
We are already changing the face of FX trading by using powerful allocation tracking methodologies built within FactSet’s core architecture. We are giving the largest buy-side players in the world access to execution methods that were never possible given their existing trading technology stack. At a high-level, the concept is actually quite simple: take a large order and make it smaller—that’s it. While this may sound like a basic thing to do, in actual practice, it’s very complicated. Tracking potentially hundreds of individual portfolio allocations and ensuring their pro-rata execution amounts are correctly accounted for when providing the fills back downstream to an OMS, in real time, is an extremely complex task. It’s a process that nearly all other trading platforms have struggled for years to solve. Fortunately, we developed our solution over four years ago and have been engineering it to be more robust and nuanced ever since; this has allowed us to make tremendous strides in delivering truly automated FX execution technology by providing curated liquidity based on trade size, currency pair, and more.
The Portware Enterprise FX module is built to offer a variety of execution styles: two-minute streaming RFQ/RFS, all-day streaming executable prices (ESPs), and various types of algo execution, both broker-based and FactSet’s in-built. These tools are not new to the industry—in fact, they’ve been around for more than 20 years (apart from algos, of course, which debuted in the late 2000s). Gone are the days where order aggregation across many portfolios would lead to an order too big to be executed electronically. The difference we provide to the market and how we’re changing the “face of FX Trading” is by allowing the largest FX trading desks in the world to access all these tools in execution sizes, which can fill their orders optimally, i.e., reducing their market impact and any associated information leakage that may otherwise result.
The traditionally high cost of trading on spreads for large orders often forces trades back onto chat or the phone, which then becomes more difficult to measure from a broker performance and TCA perspective. By allowing our clients the flexibility to break up their aggregated/netted blotter positions, they have the opportunity to look at traditional execution methods through a new and potentially automated lens.
Will FX traders be replaced by robots?
Yes, but not in the “Terminator” (or even termination) sense. Automation and Smart Order Routing are beginning to take a real hold in the FX market, and having the tools necessary to stay ahead of the trading curve will be essential to boosting any trader’s longevity on the desk. As trading desks evolve to handle workflows across multiple asset classes, the ability for a single platform to navigate them seamlessly, and to do so in a way that enables traders to spend their time on higher-value tasks, has never been in greater demand.
It’s been long discussed and highlighted that the recent trend in global multi-asset class trading is to spread the traders thin by giving them more responsibility across multiple asset classes. Whether that means FX and FI, EQ and FX, or even all the above, most desks are being forced to do more with the same or fewer bodies on the desk. This is where robots naturally come into play. A proper automation tool kit in this context relieves a trader from having to make unnecessary button clicks for routine order routing or execution; these tasks can be handled readily by the machine that is trained via the inputs provided by the expertise of the human trader. We are seeing this play out all over the world in various degrees of implementation, from simple auto-routing of WMR trades to automated competitive RFQ trades, and as advanced as automated slicing of orders and executions within curated streaming liquidity pools. It’s all there for the buy-side to utilize—they just need to come to terms with their level of comfort at having executions happen away from their hands.
What does the future of your business in APAC look like?
Trading FX is not for the faint-hearted with wild swings in the markets by everything from CPI/PPI to a tweet; the level of complexity for trading in APAC is a notch higher as the market spans across 11 time zones, consisting over 10 markets, and dozens of regional nuances and fragmented liquidity. Buy-side players in the region aren’t content with a generic trading platform, and FactSet is proud to be their technology partner providing them with a customized multi-asset trading system deeply integrated into their ecosystem, powered by a robust automation engine.
Clients in APAC were grappling with becoming a lean multi-asset class trading desk and the sudden emergence of COVID-19 has helped accelerate that process. The reinvention of the target operating model at a trading desk often begins with the consolidation of platforms and in going through this exercise, one often realizes that they are flooded with gizmos they might need but lack essential tools they must have. This has led to a real shift in consolidating platforms to focus on workflows. This challenge is amplified in APAC as trading desks possess dozens of standalone execution or broker platforms catered towards niche asset classes or illiquid markets that simply refuse to talk to one another, creating massive inefficiencies in the day-in-the-life of traders.
Our clients are demanding FactSet to challenge the status quo by not only integrating these bespoke tools into a single EMS but also offering the ability for these to be dynamically interfaced with each other and fed off central automation engines. As an example, clients expect multiple OMS to feed into a consolidated EMS that aggregates and internalizes orders before sending them to be executed by the dealing team. The traders access all asset classes and regional markets from one EMS that is fed by data from exchanges, brokers, liquidity providers, and trading venues. Furthermore, the trader is augmented by an automation engine that lays the foundation of freeing up the most critical resource of the trader—time!
How important is an investment in R&D to FactSet, and in what ways does the firm look to engage with clients to help shape your product offerings?
We partner with our clients in the development of our new products, across asset classes, but particularly in OTC markets such as FX and FI. They are partners in product development, strategic initiatives, and workflow development, all of which can be unique to them and/or scalable to the industry.
At a strategic level, we apply a pragmatic marketing framework to our approach for product development, and that means we have to thoroughly understand our clients’ (and prospective clients’) pain points on the trading desk, where they feel there is room for improvement, and where can we collaborate with them to provide innovative solutions that can be custom to the end user yet also scalable. A great recent example of this approach is with the development of our SmartRFQ framework.
The SmartRFQ module allows our clients to curate custom liquidity pools by currency pair, trade size, and execution time, while simultaneously randomizing other LPs into the RFQ pool. This feature works in concert with FactSet’s cutting-edge automation toolkit and helps clients streamline their bespoke liquidity pools while reducing potential information leakage to LPs, which might not be willing to take on certain currency risks. As an example, clients can define Tier 1 LPs by currency pair, size, and time of day; they can then define a Tier 2 pool of LPs and randomize a small subset selection from within this pool. Clients can define as many “pools” of LPs as they like and can also randomize any amount of them into a curated pool of liquidity for any given currency pair and trade size bucket.
What you think the trading room of the future will look like and how FactSet is preparing for this?
Many of us in the technology solutions business have a good view of what the trading floor of the future looks like, it’s just a matter of who can put all the relevant pieces together first; that’s who will likely win the technology race. In a big way, the future is data, metadata in fact. Enabling and empowering clients with near real-time access to historical and inflight metadata will be a massively important ingredient to how trading decisions are made in the future. This is what FactSet refers to as the “enlightened workflow.” FX TCA is now more important than it’s ever been, and although post-trade TCA is well over a decade mature in the FX industry, it’s the automated application back to the front of the trade life cycle that is becoming the new hot topic. It’s no longer the use of these cost measurements to validate previous trading decisions that are capturing desks’ attention (although that is still extremely important), it’s now about turning this data into actionable decision points for the next trade execution. Our team is creating solutions to help traders’ access historical and real-time trade data to create a curated pre-trade view into their potential best execution method. Building on the pre-trade model, clients will then be able to leverage their existing FactSet automation tools and effectively let the machine determine and execute in the most efficient size, liquidity pool, and execution method. Admittedly, we are not there yet, but we have the core building blocks in place (automation, curated liquidity, metadata storage, advanced execution tools, and in-flight TCA) to start formulating, in a very concrete way, what the trading floor of the future is going to look like.
Can you elaborate a bit more on the “enlightened workflow” and how the continuous TCA feedback loop plays into that?
Foundationally speaking, it all starts with tick data and the accompanying TCA analytics. We structure our client trading sessions in such a way that every available price tick that could be traded upon is captured within an AWS storage facility. It’s not just RFQ and execution data, we capture all-day streaming tick data at ten different levels for each currency pair that sits on the client’s trading blotter. Of course, the client can trade on any and all of the streaming prices. But even if they don’t, it’s data that is still captured at the end of the day for further analysis. Right now, it’s mostly used to compare risk transfer pricing at varying levels between a disclosed RFQ vs. a disclosed stream. FactSet has real-time tools that automate that decision for the client and execute against the best available price on either session. Stored data can reach up to 30+ gigabytes a day and if you look at it closely enough, you can start to see pricing skews within the various levels of the book and how a client’s direct credit counterparties are reacting to their executions immediately post trade.
It’s this level of analysis that becomes extremely valuable and the closer it can be calculated in real time will give trading desks the pricing transparency and edge they need to make the most “enlightened” trading decision. Having the continuous TCA feedback loop containing post-trade LP scorecards, market decay data for information leakage purposes, and curated small liquidity pools based on this data will create a real-time pre-trade decision analysis, which today, takes weeks or a month to build based on the historical TCA numbers.
Where do you expect to see your global growth coming from over the next few years with respect to providing new products and services to the FX market?
Over the next 12 months, we will continue to build upon our core technology stack, focusing on automated data-driven analytics for the front-end of the platform. We want to empower traders with referential historical and real-time data in a pre-trade staging scenario, meaning the trader will know precisely which execution tool (RFQ/Stream/Algo/Voice) will give them the best opportunity to execute their orders at the best possible price within the market with the least amount of impact. FactSet’s technology will integrate real-time market data into these pre-trade measurement tools to help facilitate the next level of automated executions.
Additionally, we are also focusing on driving further innovations by digitizing voice trading workflows to bring enhanced operational efficiencies to traders across Asia. Looking to the future, we aim to build on our successes by further expanding our asset class coverage, offering more robust workflows for price discovery and trade execution against a growing list of currencies. Having established our position as a mission-critical and state-of-the-art solution for institutional global FX trading desks, providing a sophisticated platform where automation, data-driven analytics, multi-asset capabilities, and bespoke features come together in a compelling package, will continue to be central to our focus.
Portware Enterprise is a sophisticated FX dealing platform with an established track record serving the largest asset managers globally. The platform’s design lends itself extremely well to the automation of complex workflows. Powerful allocation tracking technology built into the core architecture enables methods of execution not possible via other platforms. The platform excels at providing curated liquidity based on several trade characteristics. Its tools facilitate a variety of execution styles and can achieve meaningful cost savings over traditional methods of trading on spreads for large orders. A powerful automation toolkit gives multi-asset desks the ability to do more with less.
Traders no longer need to make a series of tedious button clicks to effect routine executions as the system can be trained to auto-route different types of orders effectively based on its users’ expertise. Data-driven analytics, automated and delivered at the point-of-trade, lie at the forefront of Portware’s R&D efforts. Traders should know which execution tool on the platform is likely to produce a preferential outcome. New products and their liquidity, along with increased regulation and further adoption of automated trading, are likely to be the main drivers of change and possible disruption in the institutional FX trading space in the near-to-medium term. Having the right toolset to take advantage of unforeseen market disruptions will be crucial to the success of any desk.