The client experience is everything and with that in mind Colin Lambert takes a look at a firm seeking to deliver analytics with a difference, as well as the impact of this service on a client.
One of the biggest changes in the foreign exchange industry in recent years has been the explosion of data and analytics products – nearly all of which come to market with the intention of improving relationships. One challenge faced by those delivering product, however, is being able to show the relevant details on a portfolio basis. Generally speaking, banks are limited to their own trades, while multi-dealer platforms are restricted to activity on their own platform – hence why so many are keen to stress why they are unable to moderate behaviour on their venues because they don’t know what is going on elsewhere.
The last couple of years have, however, also seen the rise of the independent analytics firms that can look across a portfolio of trades, platforms and providers, to paint a broader canvas of a particular firm’s activities. One such firm is FairXchange, established by former Bloomberg and Morgan Stanley e-FX executive Guy Hopkins and, as the name suggests, the firm seeks to promote a balanced relationship, based on hard data. “Our philosophy is that all participants – both liquidity providers and consumers – should be able to transact at a reasonable price, that allows both firms to make a fair return,” he says. “The data framework we have created allows that to happen by means of a thorough dialogue that can indicate both potential obstacles and opportunities to grow the relationship further.”
Part of the problem around using data and analytics to promote a healthier relationship has been that often they have been one-sided – a bank would go to their clients with their proprietary data and discuss how the relationship can be improved. When a client has several liquidity providers, all of whom bring their data in their unique proprietary format, making sense of the broader picture rapidly becomes an enormous challenge, and this is where FairXchange thinks it can make a difference. “Banks have always been very good at going to their clients with the data, but we wanted to take it further – to deliver the same data to the client in order that the discussion that takes place is on equal terms,” Hopkins says. “With our Horizon product we feel we can deliver analytics, based upon tried and tested techniques, that will help democratise the interaction and lead to more meaningful discussions. We think of it as ‘liquidity collaboration’.”
In reality, delivering independent analysis on bilateral relationships, whilst potentially challenging, is not as complex as it could be – especially if you compare it to three-way relationships such as those of a brokerage firm. Late in 2018, FairXchange partnered with Sucden Financial to deliver a liquidity management framework that allowed the firm to enhance the management of its relationships with both LPs and its clients.
“We have always talked to our clients about how they were executing, and with our LPs about the value we felt we were delivering, but we have not always been able to empirically demonstrate that,” says Wayne Roworth, co head of e-FX at Sucden Financial. “The complexity for us is that we are a buy and a sell side solution – we aim to build bespoke liquidity streams for clients and ensure they work well for both them and our LPs. This is critical to our success.
“Horizon allows us to deliver analytics that look at the issue from the client’s point of view as well as the LP’s, it’s a true three-way solution,” he adds. “It also helps us because we don’t have a large team of quants looking at this aspect on a daily basis and so it provides support to our sales and tech support teams, giving them the basis for deeper, more meaningful conversations.”
A key element from Hopkins’ perspective is the simplicity of the solution, and Roworth agrees. “It has to be a simple concept at heart, for this is still largely an education effort at multiple levels,” he observes. “It is important you are able to deliver this level of analytics, however, if you want to be recognised by your clients for more than just product delivery.”
A strong suit in the Horizon offering reflects that simplicity – the ability to anonymise the data at the click of mouse. Simply switch to anonymous mode and counterparty names that were visible are masked behind general titles such as “LP1” etc. “This really helps me in my day-to-day job,” explains Roworth. “I can look at the data and see where I believe things can be done better. It also allows me to target my conversations during interactive discussions.
“If, for example, we have an LP that has dropped off in terms of pricing quality I can look at the data and arrange to talk to them about it. The beauty is I will be looking at the same screen – the data will be presented in a familiar format while we discuss peer performance and other analytics,” he adds.
Albert Einstein once famously said, “Not everything that can be counted counts, and not everything that counts can be counted” – which could be paraphrased as, context is everything. This is no different when looking at the data thrown out by trading relationships, what looks good to one party may not look quite so rosy to the other, mainly because they are viewing it through different prisms.
In foreign exchange terms this could be explained by using the example of a corporate client that executes large tickets in a short window and has market impact, but little or no influence on longer-term direction. Looking at the flow at +10 seconds will not please the LP, however just two or three minutes later it could be worth multiple times the initial market impact to that same participant. From the corporate’s point of view, they executed in the market according to their policy, the market went against their trade, so why is the LP complaining? It is all about context.
Hopkins believes that by providing the data in context Horizon offers a real differentiator. “The context is everything,” he asserts. “It allows you to take into account different risk models, client execution style and changes in participant behaviour. It also, importantly, allows you to develop the service to best meet the needs of both you and your client.”
To this point, Horizon delivers the analytics graphically through the use of curves (the raw data can also be accessed), which give the high level view of, for example, the P&L for the business (based upon an independently sourced mid-rate), the top 10 currency pairs and liquidity providers, as well as individual client value.
The data can be viewed across time and geographical zones and value delivered in basis point, pips or dollars. Users can zoom in on the curve from -30 seconds to +20 minutes, although as Hopkins points out beyond 30 seconds post trade the data tends to get a lot noisier. He adds, “This provides a more detailed view of a participant’s business – are they doing better in Asia Pacific for example, and in what currency pairs? Where can they enhance their service in the Americas? How long should an LP ideally look to hold the risk from this counterparty?”
From the high level, users can drill down into the data to get a much more granular view, looking at attempts to trade versus actual fills, reject rates and volume traded and this can be viewed by groups of participants or by individual players. “This helps identify which LPs are strong in which pairs and what pairs offer the most value from a client flow perspective,” explains Hopkins. “Equally, we can show inception yield and the yield across the spread and how this changes by LPs, we can calculate the cost of rejections, round trip times, summarise it and present it to clients.”
The more detailed data allows Horizon to deliver context around, for example, how LPs perform differently at various times or in various markets and how this has changed. “It could be something as simple as the LP changing something or the client changing execution style,” says Hopkins. “We look at behaviour within the ecosystem, and again this comes to LP behaviour or a change in execution style; was it a new client or an LP’s hedging regime change? Whatever it is, the data will show it.”
Roworth adds, “This context is important, not only because it enables us to have more meaningful conversations, but because we can also look at the yield profiles and how they match up with different LPs. This allows us to build a liquidity stream based around what works best for the client and LP.
“By offering peer comparison – and maybe a firm’s peers have upped their game rather than that LP dropped off – it also helps us be smarter about how we allocate client flow,” he continues. “This not only allows our LPs to focus on gaps in their service, but we can also show them where they could access greater opportunities.”
By means of explanation of such opportunities, Roworth points to the yield retention curves available on Horizon. “Because Horizon can show yield retention by currency pair at a certain time this allows the LPs to make a smarter decision about how they handle the risk and what products they stream in. For Sucden, this means we can also analyse our flow as a portfolio of business to the LPs and that really helps, because, for example, we may be able to show an LP that while they are doing OK in, say EUR/USD for Client A, their pricing to other clients in NZD/USD suggests they would win more business, profitably, from this client if they also streamed a more competitive spread.”
Horizon also has sweep identification technology, although Hopkins does acknowledge that no one has really come up with a definition of sweep in the industry. “We look for clusters of trades in a short window that swamp the top of the book,” he explains. “This allows us to get into the behavioural side of the business with the clients, and help them understand how to better execute their trades without market impact.”
As this statement suggests, Hopkins is keen to stress that FairXchange is not prescriptive about what counts as good or bad flow. “We believe we have to have a flexible approach to allow the parties to look at their flow through the prism of their own models,” he says. “There are other benefits to this approach as well, for example, users can better attribute P&L within the business. We want to let people choose what they want to look at, how they look at it and what what they do with it.”
This philosophy clearly resonates with Roworth. “The ability to look at the data from multi-dimensional angles means Horizon is a very powerful tool for us that has really strengthened our business,” he says. “I feel, however, that we are just at the beginning and much more can be done as it evolves further to the extent that we can offer an even more valuable and informative liquidity service to our clients.
“It is not our philosophy to squeeze every dollar out of the LPs and our customers appreciate this approach because it underpins a more robust and stable service,” he adds. “Our customers and LPs appreciate we go to both sides and see how things can work best for both, and Horizon delivers the foundations that enable us to strengthen these relationships even further.”