While the use of direct market access (DMA) and algorithmic trading is set to see further growth among buy-side institutions in Europe, it is unlikely to approach levels experienced in the US marketplace with its relatively greater market fragmentation in trading venues and the cultural differences to Europe, a buy-side trading head has predicted.
Amid talk of the death of human traders, it is also becoming apparent that the buy-side in Europe have far more to consider than simply whether to trade through DMA or via algorithmic tools to the order books on central exchanges.
It’s not the only game in town, and sourcing liquidity on exchanges is a part of the jigsaw to ultimately fulfilling the fund managers’ requirements. In consequence, buy-side trading strategies are and will continue to be “polarised” between DMA/algo use and other channels, contends Paul Squires, deputy head of trading and securities financing at Axa Investment Managers in London.
Increased liquidity
“DMA will definitely increase post MiFID’s promulgation this November,” states Mr Squires. “It will increase as the liquidity will become fragmented and resides in more venues. An obvious way therefore to source that, is via algorithms generally and DMA to a lesser extent.”
Viewed initially as somewhat negative for its impact on sourcing liquidity, Mr Squires says he along with some of his buy-side peers believe that the Markets in Financial Instruments Directive (MiFID) will ultimately be a “small step” towards the greater transparency and the sell-side showing more of its hand to the buy-side.
But to initiate trades in large meaningful and liquid blocks with immediacy, still requires picking up the phone and speaking to a sales trader at the other end who can add some “colour” and help with discovering inventory levels on certain stocks. It’s unlikely that really substantial positions will be worked through a DMA machine due to time constraints.
“If you’re a fundamental bottom-up type fund manager where you are making decisions on whether the stock is going to be up or down 20 per cent on the year…then you’ll probably want the immediacy that you can only really get by getting more than your fair share of liquidity at any single price,” Mr Squires contends.
“What you still want from the trader is a sort of ‘inside track’ kind of comment. And, if you were really trying to discover what the inventory people have in a particular stock, it’s highly unlikely that you will discover that unless you talk by phone.”
That might be “just hold off for the buy order” despite the fact that a fund manager urgently wants a large position in a relatively illiquid stock in the lower reaches of the FTSE 100 like AB Foods.
While the influence of sales traders has been “massively diluted” in the last few years, the Axa Investment Managers deputy trading head envisages that there will be “another leg” in that since algorithms/DMA tools have been given over to the buy-side.
The trading strategy again boils down to what the particular buy-side house’s execution strategy and requirements are. “What traders have to be smarter about nowadays is exactly what they are trying to achieve in their ultimate execution objectives. And, increasingly that dictates the way you will trade.”
Reflecting on the sell-side/buy-side disconnect over DMA/algo growth, Mr Squires notes that it largely depends on who you talk to on the sell-side about penetration rates - be it agency brokers like ITG or Liquidnet or investment banks like Goldman Sachs and Merrill Lynch.
On the polarity point, Mr Squires explains that this can be separated by two poles. “Essentially, if it’s a low touch then stick the trades into an algorithm. However, if is attempting to build or dispose of a really meaningful position…then you are going to do much more block trading using a crossing network or other options. And, that is where the whole Reverse Inquiry business - similar to block trades - is so massive today.”
Banks in this space include the likes of Citigroup and Morgan Stanley. Effectively a firm with, say, a 2 per cent stake in a listed company could ask an investment bank what they would ‘bid’ them for their block in the company, on a protected discount price basis. The bank then discreetly approaches large shareholders in the company to gauge interest in the stock.
With the extremely low touch strategy, the fund manager should be able to ensure that trades “gently interact” with all the liquidity on the order books of the central exchange and a fairly ‘average’ execution is achieved (e.g. on a VWAP or TWAP strategy).
Alternative options
And, the alternatives? “If say one of our Axa fund managers has a particularly strong view on a certain stock that they want to have a critical mass position in for their portfolio(s), if I initiate that trade via an algorithm I might only secure perhaps 50,000 shares,” explains Mr Squires.
By calling a trader in the market, one of Mr Squires’ buy-side traders might be able to buy 2m shares by sourcing liquidity that is not necessarily transparent in non-displayed inventories. And, while this could cost 50 basis points more than perhaps “sticking” it through an algorithm, the extra cost is small fry if the fund manager’s hunch that the stock price shoots up 10 per cent next week comes off.
While being a cheaper option in terms of upfront basis points charged, DMA can be a “hugely time consuming” process working trades in small piecemeal blocks over the course of day. The risk of the “fat finger” trader cannot be ignored either, notes Mr Squires.
While Axa Investment Managers has an order management system (OMS) in situ, Axa Framlington does not. However, Mr Squires adds that Framlington is not prohibited from executing trades via DMA on the LSE’s order books, with a Bloomberg terminal facilitating that option.
