Miles Kumaresan, founder and CEO, Wave Labs
Nordea Asset Management’s former global head of trading is looking to revolutionise institutional bond trading as he prepares to launch a bond execution system via his start-up.
Miles Kumaresan, founder and CEO of Wave Labs, has unveiled details of the new platform exclusively to The TRADE ahead of its launch in early September.
The platform was due to launch earlier this year, but Kumaresan delayed the rollout due to the COVID-19 crisis. He has since demoed the system with six buy-side heads of trading, most of which he expects to onboard upon launch.
Known as eLiSA (Electronic Liquidity Seeking Application), the web-based platform aims to solve the complexities and challenges in institutional bond trading that Kumaresan says he faced during his time with Nordea Asset Management, and that other fixed income execution management systems (EMS) have failed to address.
“At Nordea, I really struggled to grasp the nuances of fixed income trading which were new to me at the time,” he tells The TRADE. “When the penny finally dropped and I suddenly realised the complexities of the whole process, I understood there was no way I could solve these problems from within. A fixed income EMS is a hard sell for any vendor. The systems in the market have simply replicated workflows from equities, adding no value for traders, and any adaptations in these systems have been rudimentary – simply facilitating the status quo in liquidity sourcing.”
“I knew I had to find a solution to the fixed income EMS, and in order to do so, I had to look at the fundamentals – the A, B and Cs of fixed income – and solve each of those issues. I understood that only then will the fixed income EMS truly make a difference. This is what we have spent the past two years doing, creating a system with a vast array of features that would level the playing field for all market participants – a tech suitable for buy-side, sell-side, ETF liquidity providers, hedge funds and others.”
Senior buy-side bond traders have previously made the case that most EMS platforms in the market are a tool for equities markets and not fixed income. Like Kumaresan, they argued that many of the systems have applied equities models to fixed income, despite stark differences in trading processes and liquidity landscapes.
In contrast, Kumaresan states the backbone of the eLiSA system, which was developed by Wave Labs following in-depth discussions with traders, portfolios managers and analysts, is a new workflow that is inherent to fixed income trading, focused on in-depth analysis, and execution.
“This is a very innovative piece of trading technology that has a uniquely fixed income workflow,” Christoph Hock, head of multi-asset trading at Union Investment, told The TRADE about the eLiSA platform.
The system offers multiple channels for execution, including high-touch trading, semi-automated and fully-automated execution, auto-quoting, and automated liquidity seeking and smart order negotiation and routing functionality.
eLiSA offers a suite of tools for high-touch traders for visually identifying liquidity spots and price differences, as well as a data mining tool to score, rank and make broker recommendations. Kumaresan believes a ‘black-box’ spurting out recommendations doesn’t suffice anymore. His system’s broker recommendations include a summary of why the broker is recommended, with the benefits of the recommended broker visually highlighted in comparison to all other brokers.
“This is a much more complex math problem than what meets the eye,” Kumaresan adds. “You need to have a complete lifecycle of adjusting scores up and down on a daily basis. How do you reward or penalise quotes that never resulted in a trade? How do you score two bad quotes? It’s equally challenging to give new or non-mainstream brokers a fair chance. If you don’t have much data, how can a model recommend someone with limited trade history? We believe we have a complete process for this with clear explanations.”
Elsewhere, eLiSA starts with the portfolio manager in providing granular pre-trade analysis, as well as data aggregation, historic and visualised axe, indication of interest (IOI), request for quote (RFQ) data, and the system’s flagship real-time fair value analysis.
Kumaresan explains that most bonds are interrelated and there are significant non-linearities in their relationships. This means that estimating an accurate fair value, or price, for a bond based on where others are trading is a big challenge for market participants, particularly doing so in real-time. The fair value analysis aims to solve this challenge by demonstrating recommendations that can be converted in terms of potential profit and loss.
“We take pre-trade analysis to a much more granular level in terms of what the market is doing and the fair value. It is important that the trader has confidence in the fair value recommendations as this is absolutely key to the system. It runs throughout the entire platform. This is yet another deviation from the typical equities-style use of reference quotes only in fixed income,” Kumaresan says.
The during-trade analysis also allows traders to monitor and analyse market conditions to find liquidity and track execution trends, alongside real-time estimations of probability of fill, and optimal execution parameter recommendations.
“Unlike equities, fixed income trading is closely linked to the portfolio manager and our tool starts with the portfolio manager, with pre-trade analysis and portfolio construction alternatives,” Kumaresan says.
“Then, it goes to the trader in the form of single, list, portfolio, substitute or synthetic orders. With the exception of portfolio orders, you don’t just send it to one broker or venue, you work the orders in all available liquidity pools and with many different brokers. This is realised using a variety of protocols following trader specified scheduling logic to negotiate and execute that package of bonds, to then access that highly fragmented liquidity. Wherever the other side of a trade is posted, eLiSA will find it.”
For execution, eLiSA provides several order types including portfolio trading, which has surged in popularity with buy-side traders recently. However, Kumaresan explains that when engaging with portfolio trades, it can be difficult to know if the price of a bond from the 200 names in a portfolio is, in fact, a good price. His platform aims to offer more advanced slippage decomposition analysis and optimal execution evaluation to solve this challenge, with the aim of reducing transaction costs.
“With portfolio trading, it’s extremely important to get that price back from the broker, decompose it and see for yourself if you are being priced fairly or not on a bond-by-bond basis. You may see that 90% of the pricing is indeed fair, and want to remove the outlier bonds from the portfolio and source it separately. This is another way in which traders add significant value.
“In a world where it is very difficult to know where a bond should be priced, verifying the broker pricing of hundreds of names in a portfolio in addition to all other trading tasks is a daunting prospect without tools to assist. But with our system, from a compliance point of view, at the click of a button one can document best execution in large portfolio trades,” Kumaresan says.
“The much-needed paradigm shift in fixed income trading needs catalysts for change and eLiSA may just be one of them. While liquidity in fixed income is highly fragmented, there is plenty of liquidity to be found at the right price and place. The challenge is for the holders of liquidity to make this available. Imagine a marketplace in which the natural holders of liquidity, namely asset managers, were to start offering portions of their liquidity? This would revolutionise the 1970s style market structure that we still have now in 2020.”
The buy-side has increasingly leaned towards taking on a price maker role in fixed income markets as relationships with dealers have evolved. BlackRock’s head of global trading, Supurna VedBrat, has previously highlighted that she sees the buy-side taking on the alternative role in disclosing prices, which could be beneficial for asset managers and the liquidity landscape.
Kumaresan adds the price making functionality on the eLiSA system offers any asset manager the ability to make prices in thousands of bonds while satisfying portfolio constraints. Using an example of a typical fund having positions in a smaller subset of names making up the benchmark index it tracks, Kumaresan explains that a position in 500 names out of 2,500 leaves the trader synthetically short of 2,000. By making prices on the 2,000 bonds, Kumaresan says, the fund achieves a reduction in tracking error and can capture bid/ask spreads associated with the illiquid bonds.
“This activity alone can boost the performance of the fund by 25bps or more if done right – meaning, if you usually make 50bps annualised, now you are making 75bps or more. The net effect of this is the fund’s performance ranking will climb upwards,” Kumaresan claims. “This is particularly useful in extreme or volatile market conditions, where besides capturing much larger gains, traders can help stabilise the market. A natural consequence of the price making activity is the flooding of new liquidity into electronic venues. Even illiquid venues can attract liquidity, so long as they offer innovative protocols.”
The execution platform is currently connected to a major fixed income trading venue, with plans to add more in the near future. It also includes auto-negotiating functionality which aims to give traders the opportunity to work multiple venues or multiple orders at the same time with zero touch, according to strict conditions set by the user.
“The auto-negotiation is like cruise control, and it’s extremely useful if you’re the driver with so many other things happening. It is just not practically possible for a trader to simultaneously work 300 names, and that’s why so many opt for RFQ because finding liquidity is so difficult and time-consuming at venues. The trader should do the thinking part and let the technology do the laborious liquidity seeking part.”
“Other systems simply operate as an interface to send orders to other platforms to be worked from there, which is a very manual process. With the automated liquidity seeking tool, our platform will find the liquidity, negotiate multiple bonds for you, and help navigate the execution.”