1. To what extent have new regulations and the move towards Best Execution regimes created interest in algorithmic FX trading solutions amongst your Canadian clients?
The arrival of new regulation is by no means the only factor driving algo adoption but, certainly we see clients using algos as a means of achieving their best execution objectives.
In RBC’s view, algo trading is a vital component of a wider offering which serves electronic and voice orders and RFQs, alongside benchmark execution. Clients will select which route to take depending on different prevailing factors. What’s important is that we have each base covered. Our duty is to inform clients when an algorithmic approach will provide the best possible outcome.
2. Canada boasts a large and diverse community of buy-side players, many of whom are technologically sophisticated. Which strategies are currently proving most popular with them?
Canada’s buyside has been keen for us to customize liquidity pools as more sophisticated clients are acutely aware of the impact of trading USD/CAD, a liquid currency with a relatively small ADV. For clients preferring DMA, RBC has simplified the execution strategies accordingly, providing fine-grained control over price and destination routing.
Having been somewhat wary during the first wave of algo adoption, Canada’s buyside is now demanding sophisticated and mindful execution rather than accepting generic liquidity and default behaviours from their execution partners.
3. How has RBC tried to shape the future landscape of algo trading?
RBC was involved in the initial design and architecture development of the FX Global Code. RBC led working groups that drafted the Electronic Trading and Confirmation & Settlement sections of the Code. We remain a strong advocate of the Code and signed the Statement of Commitment in February, 2018. We recognize the Code as good practice in wholesale foreign exchange markets and have committed ourselves to conducting business in a manner consistent with the Code’s principles. Earlier this year, RBC published a FICC Client Notice restating some of the key terms of our trading relationship with clients. We continue to shape the landscape and direction of the FX industry via CFEC and the Bank of England FX Joint Standing Committee (FX JSC).
4. What do you see as the key value propositions of FX algo execution and how are you encouraging RBC clients to adopt it?
Assuming a quality algo that is executed well, the key benefit to the end user will be transparency.
5. How do you ensure you avoid any potential conflicts of interest with your clients utilising FX algos?
Transparency is a recurring theme here. We are transparent about where we route trades and how we trade. This data is available to RBC clients. We offer a choice of execution models including full arm’s length execution which leverages RBC’s 24-hr Futures desk. Here, FX algo execution is segregated from the FX front-office. To prevent other potential conflicts, our trading desk has no visibility of any algo flow, pre or post-trade. For a number of clients we also use segregated Bloomberg chats to provide pre/post-trade execution analytics.
6. In what ways are your quantitative trading development teams starting to use new technologies like AI and Machine Learning in order to build a new generation of smarter FX execution algos?
In 2016 RBC established Borealis AI, a Canada-based institute dedicated to state-of-the-art research in machine learning and artificial intelligence. Working together with Borealis, the team at RBC has invested heavily to apply Deep Reinforcement Learning to improve trading for the benefit of RBC clients. Together, the teams at RBC and Borealis AI have made major progress on real-world applications for reinforcement learning. The future for RBC’s FX algo clients looks very bright indeed.
7. How much demand is there from Canadian buyside firms for a wider spectrum of FX execution algos (eg. Non SpotFX)?
There is strong demand across corporations in Canada for Latam NDF currency solutions and naturally we expect to see demand shift from pure RFQ to algos. RBC is currently developing solutions to provide NDF algos. RBC’s clients already benefit from workflow solutions which allow executions to be rolled forward and allocated pre or post-trade.
8. What has RBC done to avoid the stigma of “black-box technology”?
As we’ve mentioned, transparency is a key and it is this that will help erase any such stigma. The TCA for placed and filled orders openly reveals the algo’s logic throughout the execution process and all of the underlying assumptions can be explained. Furthermore, we trade algos here at RBC. And we would never put something in front of a client that we wouldn’t use ourselves.
9. How are you working to enhance your FX algo toolsets?
In the very beginning, our product build included a highly transparent TCA client solution and we continue to evolve the presentation of this data and execution benchmarks. We see increasing demand for integration into third party TCA solutions – something we support already. Third party tools are becoming more demanding in meta-data tagging of orders to help explain execution outcomes. We are partnering with a number of firms to provide this where required.
10. Does the Canadian FX market have any special regional characteristics that are likely to influence the traditional reasons for using FX algos?
Understanding the market structure of the currencies in which you trade is the key component of designing a successful algo strategy. RBC understands the USD/CAD market very well, and we have access to unique liquidity as a major market-maker. Our role has always been to partner with clients and offer them the benefits of working with a strong CAD bank. We believe the transparency of our products has built a bedrock of client trust and our flexible, dependable approach to execution has earned RBC client loyalty.
