Axa moves to EMS for granularity in execution control

By:
Published on:

In seeking the most efficient route to market, Axa Investment Managers has made the decision to focus on new trading protocols on an execution management system (EMS).

published in association with:
thomson reuters  
•Beware a liquidity crunch in FX
•E-trading good but not necessarily best execution
•RMB two-way volatility a key hedging challenge
•Price to pay for liquidity in technology arms race
•The return of relationship banking
Features 
•FX probe digs the dirt to clean up market structure
•The impact of regulatory change on corporate FX management
•The effects of volatility on liquidity and fixed income

cover
Return to the supplement index

Download full version of guide (PDF) 
The thinking behind the move was to create greater flexibility, says Lee Sanders, Axa IM’s London-based head of FX/MM and UK and Asia fixed income trading.

"In FX we decided to move away from the established ECN market to an EMS, which gives a lot more access to innovation and versatility in the market," Sanders says. "It means we can hook into the same system and take prices from any of the 15 banks with which we want to deal, which is better from our perspective than the request for quote protocol that dominated previously."

The difference between OMS and EMS is that the latter enables direct market access, as opposed to facilitating trades through an intermediary, and while OMS excels at portfolio accounting, analytics, and compliance, EMS offers higher levels of  granularity in execution control. An example is in algo trading, Sanders says.

"We can use the EMS to connect to anyone’s algo, releasing it from the staging area on the EMS straight to a bank’s algo suite, rather than releasing via the OMS. I can send an RFQ order, resting order, fixed order or limit order; so I get all of that versatility, depending on the counterparties’ API."

AXA IM will be looking at tailor-making its trading algos, working with banks, Sanders says. 

The firm executes around 75% of its tickets electronically, and the rest by voice. However, voice still accounts for 40% of volumes, as bigger tickets are more often executed on the phone.

Still, EMS brings with it some key advantages, Sanders says, including its anonymity and its ability to select execution venues, including algos.

"We can use the algo to execute in a way that reflects the particular investment driver," Sanders says. "If we want to be aggressive we can keep lifting the offers until we are done, or we can be more selective and passive and sit back and be the offer or the bid, which may potentially give us a better fill. On the other hand if it’s one minute before non-farm payrolls you can go all out with a big order and while you might get some slippage you should be able to get it done.

"The other thing we like about our EMS use is as a clearer and in providing better transaction cost analysis. This is not just a box-ticking exercise for us; it’s a post-trade functionality that goes a long way to help us gain more efficiency and has proved useful for our pre trade analysis."

AXA IM uses TCA to evaluate its trading performance against three benchmarks: the WM/Reuters foreign exchange fixing, creation time and execution time.

"Those are what we use to analyse where we are adding value. For example would it be better doing everything at the (WM) fix, or are we losing value in the time it takes to get from the fund manager’s request to executing in the market, or at execution are we getting the best price possible?

Use of TCA functionality has had real impacts on the AXA IM business. 

"We run through all the currency pairs and see if there is anything we can do better, and we analyse all the funds and the trades that have been executed on their behalf. We look to identify weakness caused by order creation time, lack of documentation and risk constraints, executing away from prime liquidity points and lack of liquidity through restricted execution venues.

AXA IM also uses TCA to evaluate counterparties, creating peer groups and calculating average levels of 'added value’ before pinpointing where each counterparty’s strengths or weaknesses lie.