Leveraging algos as part of a market leading FX execution service

With Richard Turner, Senior Trader, Currency Solutions at Insight Investment.

Insight Investment is a leading asset manager focused on designing investment solutions to meet its clients’ needs. Founded in 2002, Insight’s collaborative approach has delivered both investment performance and growth in assets under management. Insight managed £836bn1 of assets as at 30 September 2021 across its core liability-driven investment, risk management, full-spectrum fixed income, currency and absolute return capabilities. We asked Rich Turner to tell more about how the currency trading team of this large global firm harnesses the power of algorithmic trading.

Richard Turner
Richard Turner

Richard, please give us a little background on your career to date and how you became involved in FX?

When I left school, I started a student apprenticeship in engineering. However on finishing my engineering degree I moved into commodity trading for a large private firm. From commodities I moved into trading FX and interest rates whereupon I moved to Pareto Partners (now Insight Investment) to specialise in FX.

What do your typical day to day responsibilities within Insight Investment involve and how has your job evolved over the past few years?

Insight Investment by numbers

We execute all the FX requirements of the many stakeholders who need our services at Insight. We trade all FX products (Spot/Fwd/Swaps/Futures/Options) and we aim to deliver “best in class” FX services in all products. Over the years the Best Execution policy has evolved, and we have strived to improve the method in which we execute all aspects of FX to reflect this laterally incorporating the FX Global Code. Over the years the method of execution has become more electronic. With the transparency that this offers, we have increasingly analysed the methods in which we trade to try and create a feedback loop so that we are better informed in the future. In essence we now spend a lot more time analysing the various pieces of data that we receive on execution to better inform our “point of trade” decision making.

What do you see as the most important attributes that a good FX trader needs to have nowadays?

Insight Currency Absolute Return Exposure

The ability to evolve. Over the years the modus operandi of a trader has changed massively. The electronification of the market and all the data that is captured on the back of trading in this way has meant that traders now spend a lot of time analysing data. A lot of time is actually spent looking through the reams of trading data to better inform future trading decisions.

Traders now rely a lot on technology to supplement this and so an understanding of technology is also becoming an increasingly important factor of trading roles. One pattern that is certainly evolving is the ability of the trader to utilise programming skills to deliver reports from the multitudes of information that lands on his/her desk. Today, traders are required to code in Python (and the like) as part of their day-to-day job. I think this may well be increasingly required in the years ahead.

What are the main objectives of your currency trading team and in what ways does its execution expertise contribute and compliment to the overall investment process?

One of our team aims is to deliver best execution regulatory requirements whilst minimising market footprint. We also adhere to the behaviours and expectations of the FX Global Code. Specifically, we aim to deliver the best possible execution outcomes for all of our different stakeholders. We curate liquidity pools and utilise bank algorithms that will deliver the best outcome for the differing requirements of our stakeholders which is additive to the investment process.

How much emphasis do you place on continually working to improve your technology stack and trading platforms?

Utilising technology is becoming an ever-increasing part of our day. Improving trade workflows by utilising technology is fast moving the trader of the future into a job akin to an airplane pilot flying on autopilot. The trader of the future will allow technology to automate the trades that do not need attention (with appropriate Transaction Cost Analysis (TCA)), allowing the trader to focus on the trades where he can add distinctive value to the process. At Insight we continually look to improve our trading outcomes to compliment the overall investment process. Technology improvements form a large portion of this process.

Traders now spend a lot of time analysing data

When did Insight Investment start to explore the use of FX algorithms and how difficult has it been to integrate them into your trading operations?

At Insight we have been trading FX electronically for many years. Initially we utilised “click and deal” to execute, which quickly evolved into bank algorithms. We were first users of many of the bank algorithms and have been on this journey to electronification from the very beginning. Complimenting this approach, we have continually improved our trading operations to allow us to collect and use the data that comes with electronification. We have also looked to strategically partner with certain vendors to deliver best in breed TCA. We are constantly looking at ways to automate our many processes.

What direct benefits does the use of trade execution algorithms deliver for you?

Transparency of execution, adherence to best execution requirements as well as adherence to the principals of the FX Global Code.
What factors might influence the sort of FX algorithms you employ and the type of orders you work with them?

We do a lot of work around behaviours of child order fills within algorithms, as well as overall market slippage. We look to analyse the outcomes to ensure that they meet with the expectations we have at execution onset.

Underlying market conditions are also very important to us and our method of execution is often driven by how the market is behaving.

Insight has been trading FX electronically for many years

What are you particularly looking for in terms of the functionality that FX algos can offer?

We utilise algorithms that provide in-flight transparency around execution. We also make sure that the behaviours of the algorithms match our expectations of them. The FX Global Code has brought focus onto the behaviours of execution algorithms (as well as other methods of trading) and we curate algorithms that satisfy adherence to the FX Global Code.

To what extent does an understanding of market dynamics and liquidity influence the decisions you take on how to deploy an FX algorithm and the execution style that may be most appropriate?

An understanding of the underlying market conditions at time of trade is vitally important. Different currency pairs have very different (and sometime similar!) profiles. If you also consider the nuances of NDF currencies the question of where/how to trade can be very challenging. At Insight we utilise as much data as we can to inform our decision at point of trade. In some instances, given tricky market conditions, the decision to risk transfer the entire execution via RFQ may provide us with the best outcome.

In other instances, it may prove best to remain very passive and take some time risk. The problem we sometimes encounter is that, ultimately, until you are actually “in the trade” you can’t say with 100% confidence what the market conditions are. This is an area we are looking to improve in the forthcoming years,

How “hands-off” are you prepared to be once you have committed to using an algo or do you still like to micro-manage various parts of the execution process or at least constantly monitor it?

The members of the trading team at Insight have very different approaches to execution. We encourage the team to utilise the data as much as possible to inform their decision of which algos/liquidity pools to use. Ultimately however, different traders have different styles and have differing approaches to execution management. Some frequently trigger limits whilst others may toggle liquidity parameters such as urgency. Ultimately these interactions can be analysed using TCA to see the trader’s alpha.

Have you been able to establish an algorithmic FX trading policy to deliver evidential results that can prove that your method of execution with these toolsets is robust and consistently good?

Yes. Over the years we can evidence this utilising our TCA. We use several benchmarks to do this but ultimately our execution performance is evolving with the market and the underlying liquidity conditions. We are constantly analysing our data for better ways to trade.

The problem we sometimes encounter is that, ultimately, until you are actually “in the trade” you can’t say with 100% confidence what the market conditions are

How do you source the FX algorithms you use and what sort of information about their capabilities and performance attributes are you seeking from providers?

We have an algorithmic trading policy which governs the way we choose algorithms. We also have regular meetings (as well as live discussions) to look at changes in the methods we employ to execute trades. We look to our independent TCA providers to help us analyse the data that is sent to us from our counterparty banks. We also challenge our counterparties to give us standardised data so that we can be homogenous in our data analytics. Data is a key point, and we are constantly asking our counterparties for more data where we think we need it. How has the way you work with FX algos changed since you first started using them? For example are you trying to make their selection more systematic and are you prepared to experiment with relatively unproven and brand new algos.

Our Algorithmic Trading Policy enables us to put structure around the algos that we use. Our policy is constantly evolving as the liquidity landscape changes. For Insight, liquidity provision is key and we try to ensure that the liquidity we use is consistent with out Best Execution policy, as well as adhering to the FX Global Code. If we utilise a new algorithm there is a very deep dive into the algorithm’s design and whether it is suitable and consistent to our trading policy. Once we are comfortable with this, we then forensically monitor the algo in-flight. There is then a post trade deep dive into the performance of the algo. A decision on its suitability is then made. Ultimately all the algos we use are being constantly monitored for their performance and suitability to our flow.

How much has the Covid-19 pandemic and working from home environment acted as catalyst for your team to increase their level of trading automation including the use of algos?

Before the onset of Covid-19, we were already extensively utilising algorithms. The pandemic only served to cement the idea that what we were doing was the correct approach. We had some positive outcomes over the very volatile periods. The speed to market and adaptive nature of our electronic execution performed very well.

Looking forward we are aiming to utilise rules-based trading to allow a degree of automation in our trading. We will be looking to utilise automation to reduce the number of manual tickets and allow us to concentrate our efforts on the trades which are trickier in nature. We have an exceptions framework to capture any outliers and highlight any issues.

Today, traders are required to code in Python (and the like) as part of their day-to-day job

Some firms rely on their own internal TCA toolsets to better inform their decision-making process at the point of trade execution whilst others work with external independent third-party TCA providers. How do you approach this task?

We have external TCA providers to provide our clients analysis which is independent. We utilise this independence to meet our client requests for execution information such as costs. However, we do perform our own TCA when required. Often, we will do deep dives into execution analysis and this is performed utilising BOTH third party TCA providers and our own internal TCA. We are constantly challenging our approach to execution.

How far can leading buyside firms like yours go with algorithmic trading? In other words is the ultimate objective fully no-touch execution or is that going too far?

Algorithmic trading gives very transparent execution. With any approach that is not electronic, this transparency is potentially lost, as subjective information must be used. At Insight we believe that we should be as transparent as possible around execution. We will look to increase our degree of no-touch execution where we can. However, we are realistic in our approach to realise there are times when we must utilise manual execution. I firmly believe that over time our percentage of no-touch execution will increase but we will still be needed to have some inputs into the trickier trades. I would also expect this to happen over asset classes other than FX.

You have talked in the past about wanting to see the industry improve its transparency around rejection codes, type of liquidity, cross currency liquidity and last look. Are you satisfied with the progress that’s been made so far on these issues?

Transparency and consistency of data is the key issue here. The “Holy FX data Grail” is the aim, where all algorithmic counterparties supply the data mentioned in a standardised and timely fashion. I was involved in a lot of work with The FX Global Code around TCA and data and the focus was very much on improving standards. There is now a template that can be downloaded and utilised, and our hope is that, ultimately, everyone uses the same format to send data.

This approach enables us to challenge the counterparties to fill in the gaps where we see inconsistencies across the data set. Insight was also involved with the Global FX Committee’s Disclosures Working Group, led by the Bank of England. We will be encouraging all our liquidity providers to complete these disclosure templates and make them publicly available for their liquidity consumers. We use this alongside TCA data and the completeness of data mentioned above, to rate our counterparties at reviews.

In what ways do you think the increasing use of sophisticated yet flexible toolsets like FX algos will change the role of the FX trader in the future?

The role of the FX trader is constantly evolving, and the increased use of electronification/rules-based trading will enable the trader to concentrate more effort on trading that requires attention. Manual tasks will be automated, removing the potential for key-stroke risks. This will enable the trader of tomorrow to focus more attention on data analysis. Managing counterparty relationships will become more transparent and attention will be focused further on execution performance and costs. There is still a lot of work to be done regarding transparency of trading costs across ALL asset classes.

At Insight we pride ourselves in constantly challenging ourselves to do better

Looking ahead how ambitious are your plans to leverage the power of FX algos still further and do you expect them to play an increasingly important role at Insight Investment?

FX algos give transparency to execution and allow us to meet our best execution requirements whilst adhering to The FX Global Code. We can do a tremendous amount of analysis, with electronic trading, which allows us to forensically improve our performance over time. I would hope that we will increase our usage of FX algos going forward but we will only do so when our analysis suggests that is the right thing to do. At Insight we pride ourselves in constantly challenging ourselves to do better. I hope to see this continue by utilising new developments in FX algorithms amongst other suitable technology solutions.

1 As at 30 June 2021. Assets under management (AUM) are represented by the value of cash securities and other economic exposure managed for clients. Figures shown in GBP. Reflects the AUM of Insight, the corporate brand for certain companies operated by Insight Investment Management Limited (IIML). Insight includes, among others, Insight Investment Management (Global) Limited (IIMG), Insight Investment International Limited (IIIL), Insight Investment Management (Europe) Limited (IIMEL) and Insight North America LLC (INA), each of which provides asset management services.

Related Posts

Scroll to Top