Dmitry Ilyaev
Dmitry Ilyaev

Exploring the shifting ecology of FX Liquidity and future role of banks as providers.

e-Forex talks to Dmitry Ilyaev, Head of eFX and Spot Trading at Commerzbank

A recent poll found that liquidity concerns have taken over from best execution requirements as being the greatest concern for FX traders. Are you surprised at that and why is the FX liquidity that the investment community requires not always there?

FX is an opaque, decentralised market prone to information asymmetry between participants. Visible liquidity is not necessarily dealable, and even when it is, there are other considerations outside of ‘best price’, such as market impact and opportunity cost of rejections, that a liquidity consumer may need to consider in order to reduce their execution costs. The proliferation of liquidity providers which are not holding risk means that a liquidity consumer who is not asking the right questions when constructing their liquidity stack may find it difficult to fulfil their liquidity requirements and ultimately pay a higher price to transact FX.

In what ways is FX liquidity provision becoming more challenging for principal market makers?

The industry trend is towards reduced ticket sizes and algorithmic execution. Principal market makers therefore find themselves fighting for client flow in an environment of intense competition and diminishing margins. Against this backdrop, it is important for a market maker to make ongoing investments to their technology, IP, low latency networks and staff in order to remain relevant.

In what ways is the changing FX Liquidity ecosystem forcing buy-side firms to adapt their trading strategies to achieve best execution?

Buy side firms looking to control their FX execution costs increasingly have to engage with their liquidity providers in a more sophisticated way. This means adopting a more data driven approach to decision making, both for optimising execution strategy parameters and for deciding on which liquidity providers to include in their liquidity stack. For clients utilising third party execution algos, it is important to quantitatively evaluate the performance of each provider’s algos against an objective benchmark, controlling the results for prevailing market conditions at the time of the execution.

Could we start to see new trading models emerge to try and solve the problem of shallow liquidity? For example with a form of liquidity aggregation by extending the time window for liquidity sourcing?

Boosting access to liquidity by simply increasing the number of liquidity sources available to the execution strategy can be challenging. Including liquidity providers who don’t have the scale to warehouse risk can amplify market impact and discourage genuine providers from showing competitive prices. Extending the time window of execution is one way to alleviate potential market impact, but this will always expose the execution to higher levels of market risk. One potential solution is to source liquidity from correlated products, which would allow the participant to slow
down the execution in the target market while controlling the overall portfolio market risk.

The onus is on market participants to conduct their own due diligence when cultivating their pool of liquidity for trading
The onus is on market participants to conduct their own due diligence when cultivating their pool of liquidity for trading

Why have many buy-side firms, especially those moving large orders, become more sensitive to the concept of depth-of-book liquidity rather than just top-of-book liquidity?

Many buy side firms are justifiably wary of information leakage through transacting numerous smaller child orders at the top of the book. As a way of addressing this, some choose to deal in size through several layers of liquidity. To combat the steady decline of liquidity in the primary venues, these dealers prefer to build up their own custom pools of liquidity made up of various non-firm ECNs and direct API LPs in order to boost available liquidity at depth. For this to work effectively, dealers must ensure the constituent sources are independent. A few large LPs may dominate liquidity provision across multiple venues, and seemingly independent LPs may use the same sources to construct their prices, so trading with one of them may signal that information to others. In such a complex, interconnected ecosystem, a deep order book can sometimes be a mirage, with duplicated non-firm liquidity often leading to disappointing fill rates and excessive market impact.

FX remains as fragmented as ever and the quality of trading venues varies as does their liquidity profiles. How much of a challenge is it for banks and clients alike to benchmark the various liquidity pools available?

Over the last few years, venues have started to acknowledge that to achieve sustainable growth they need to look beyond volumes and encourage responsible trading behaviour from all participants. Conversations between the venue operators and market participants have shifted from top of book spreads to fill rates and now to minimising market impact and information leakage.

This is a positive development for the industry. However, the onus is on market participants to conduct their own due diligence when cultivating their pool of liquidity for trading. Important decisions need to be made around choice of venue, mix of LPs on those venues, location of the venues relative to their execution engine, etc. Objective metrics around the opportunity cost of rejections, market impact and prices need to drive order routing decisions.

As the buy side looks for smarter ways to navigate the fast changing FX liquidity environment how much growth in the use of execution algos could result from this and what key advantages do these toolsets have?

Execution algos can be a double-edged sword. While they offer clients greater control over their executions and the resulting TCA brings about much needed transparency, clients should also be mindful of the potential pitfalls. The client is exposed to market risk throughout the duration of the execution, and misinformed decisions around algo parameters can bring about suboptimal results, such as incomplete order fills or excessive market impact. It is the role of the algo provider to guide their clients in making well informed decisions.

Buyside firms are becoming more versed on the impact that good or bad liquidity can have on their executions. How are leading FX practitioners responding to this by helping them to measure and better predict the cost of liquidity?

Ongoing dialogue between LPs and their clients is vital. A good provider will be less focused on selling their clients a black box solution and more focused on educating them on the potential pitfalls of liquidity mismanagement. LPs need to be willing to share their expertise on topics such as liquidity profiling and market impact, and foster a data driven approach to decision making. By arming the client with the right tools, it is the numbers that will do the selling.