By Fabien Oreve, Global Head of Trading, Candriam
Market fragmentation and tougher trade execution reporting requirements compel greater investment in technology and also a more stringent selection of brokers.
A major intention of the Markets in Financial Instruments Directive (MiFID) II is to improve the transparency of financial markets, and hence ensure greater justification of best execution for each major asset class. This dual objective of transparency and justification is not one that market participants always find easy to meet; one of the biggest obstacles is growing market fragmentation and another, the degree of technological investment required.
Managing market fragmentation
The Candriam trading desk began by revamping its procedures, strengthening order-execution processes by providing greater detail and case studies. It designed a more detailed execution policy with decision trees to identify best execution from pre-trade to post-trade. For example, we now describe in greater detail how pre-trade indicators feed into the broker-selection and best-execution processes. Data-driven decision-making is a key aspect of this evolution.
The desk streamlined its lists of brokers and platform providers in order to focus on key partners who know our business inside out. Naturally broker lists have been reduced, in particular for equities. In each business category, we have retained a limited and balanced number of brokers, which entails excluding those we perceive as having no added value. With more reporting obligations and justification exercises, we have tried to streamline connections, focusing on “one-stop shop” brokers and platforms that offer easy access to liquidity. The core list of brokers has been complemented by a group of specialists to cater for more specific business activities.
In keeping with the spirit of MiFID II, amid greater market fragmentation, the desk has tried to gain agility in the way it manages equity orders across markets through dynamic multi-placement. Our order and execution management system (OEMS) provides a functionality to split a single order by the number of venues and brokers it has been placed with. We can take advantage of any potential increased liquidity through the OEMS. For example, the original order placed with broker “A” on lit venues is automatically reduced in our system if a portion of the order is matched with a natural block offered by broker “B”.
With the return of market volatility, we have recently been trading more in emergency mode and relied on systematic internalisers (SIs) to reduce the timing risk. We tend to focus on our large brokers, who facilitate trading for orders in their entirety. The selected brokers are typically those who have made major upgrades to their central risk books, where they hedge positions more efficiently.
It’s uncertain as to which trading venues will benefit most from the new MiFID II rules, for instance, from dark pool restrictions, but agility for a buy-side dealing desk in executing orders across brokers and venues will certainly gain ground thanks to the OEMS.
New regulations have clearly led asset managers in general to make significant adjustments. These include upgrading their technology and gain efficiency in dealing, enhancing monitoring capabilities, sending legal entity identifiers to brokers and platforms, collecting more trade data and ensuring a more systematic approach to best execution.
We have also had to work on further automating our order dealing process in asset classes other than equities without compromising security or order execution quality. FIX connectivity, request-for-quote (RFQ) platforms for bonds, currencies and exchange traded funds as well as algorithmic trading for equities have already, to a certain extent, automated our trading desk.
However, we have never had full-scope automation as an option. Why? Because, for example, at the Candriam trading desk, sending an equity order to an algorithm is subject to strict conditions and written agreements with brokers on pre-trade constraints like maximum order value and maximum average daily volume percentages.
With today’s OEMSs, there is more room for automation and more opportunities to further streamline our trades. That is true for a portion of our futures and FX order flow. We have selected a multi-asset “algo” broker covering equities, futures and FX, and are currently working with the IT teams, legal, compliance and risk management departments to set up trading limits and to document and test new tools.
FX is an interesting large asset class where a majority of our flow is handled via electronic RFQs. For the most liquid currency pairs, our RFQ platform can be clearly complemented by other tools, such as time-weighted average price (TWAP) and by time- or price-trigger algorithms (equity-like trading mechanisms) to meet portfolio managers’ requirements. However, we do not see any interest in using electronic RFQs for cash equity, because we currently have easy access to indications of interest (IOIs) from those of our brokers who advertise reliable tradable sizes and prices.
As a result of MIFID II, the main challenge with transaction cost analysis (TCA) lies in assessing and ranking brokers in a more rigorous, quantitative and yet non-complex fashion. Ranking brokers that have performed under similar scenarios is the first step to more easily identifying outperformers versus underperformers.
Another challenge is that of presenting meaningful post-trade data across the major asset classes: that includes fixed-income, which requires more manual analysis and verification than equities.
Candriam has been engaged in post-trade analysis for equities and fixed-income for a number of years. In the case of fixed-income, we internally built a tool that produces reports where each trade is compared against the losing quotes obtained from the RFQ process and the “composite” mid-price level at the time of execution. Of course, there are still market data issues for the most illiquid bonds, and this does not help transaction cost analysis but, we have managed to find a way to know our trading costs, business trends and outliers across different bond segments and sub-segments.
For any given bond category, our post-trade reports provide average bid-ask spreads and executed prices’ deviations versus. mid-price reference. These reports also help visualize top broker-dealers. Knowing top broker-dealers in specific segments is quite useful for fixed-income traders and helps them target counterparties.
As the scope for TCA has become far broader under MIFID II, we are also discussing potential new partnerships with multi-asset TCA providers covering equities, fixed income and FXs
Impact on trading desk organization
MiFID II promotes market transparency and greater convergence among asset classes. The new regulations are driving more order flows to electronic trading. While electronic trading technology is broadly used today, traditional trading methods still have a significant role to play. At the Candriam trading desk, traders have to work more horizontally and navigate between low-touch and high-touch trading channels, depending on order difficulty and market environment.
Having junior colleagues’ roles confined to low-touch trading only is not a satisfactory solution. Bringing together people with various backgrounds and diverse perspectives, training new employees to find the right balance between electronic and voice trading for specific orders in a particular asset class enhances professional growth and expertise.
Trading desks always need the flexibility of human traders to adapt to changing market conditions and achieve best execution. After all: “It is men who make a city, not walls or ships” (Thucydides).