Developing Trading Algorithms – Buy Vs Build

Electronic Trading Infrastructure & Algorithms

It is always hard when you hear the phrase ‘veteran of the industry’, especially when you realise they are talking about you. However, given the youth and innovation that exists in electronic trading, I find it hard to avoid using the term about myself. I started my career in 1983 working on the London Stock Exchanges first generation of electronic trading infrastructure, information systems, and big bang One (the move to an order driven market – one of the earliest implementations of screen-based trading).

Buy Vs Build

One of the perennial questions that I have encountered over the years is the debate over Buy Vs Build. I once heard that if the IT departments of many investment banks were standalone tech companies they would be large enough to be in the FT100. Whether or not this is true it serves to illustrate the sheer size of the big financial firm’s IT departments, employing thousands of some of the best technology brains on the planet. So there is some inevitability that the first place to look for a technology build will be within your own firm. But does this really make sense in an era of the need to reduce costs and take a more holistic view of the Total Cost of ownership.

So, the buy decision can be hard. It takes a strong pragmatic leader to go down this route where the heart says no, but the head says yes. Surely it is better to work with a strong partner with a proven track record. That way you take away the development and implementation risks and to have technology that is continually innovating on someone else’s time.

At Tradetech in Paris in 2017, they held an Oxford style debate about this very subject. Remembering that the skill in an Oxford style debate is not about what the parties believe, but how well they argue the case. We learnt from the debate there was a level of consensus that firms can do most things in the time-frames. However, when weighed up with the cost of build, the resources and time available often make it difficult. The challenge for vendors therefore is pricing services at a level that make them attractive under those circumstances. However, vendors do have an advantage in non-differentiating solutions. They can amortise the development costs across multiple customers, where in-house IT normally only have one customer.

Software Houses

On the other foot, however, are the software houses, whose life blood is about providing software to companies. Whilst many of them offer shrink-wrapped products that firms can simply slide into their business model, many houses provide complex infrastructure to complement in-house builds. For these firms, their one vision must be to deliver complementary, easy-to-deploy great quality software

In the same way, it is a challenge for software houses to keep an eye on what they are good at and not constantly wanting to increase their boundaries and expand their product offering such that they become master of none.

Having moved from the sell side into the vendor community, I have found it comforting that there are still firms such as RA that believe in the philosophy that you should understand what you are good at and stick with it.

Developing Trading Algorithms

Developing trading algorithms is a case in point. RA does not provide Algorithmic trading technology per se. In order to differentiate algorithmic trading services, sell side firms need to keep control of this part of the technology. Under MiFID II it is no longer sufficient to provide a good service in one part of the business and be paid in another. Execution quality must stand alone.

What RA does, however, is to provide the technical environment on which firms are able to build their own intellectual property without the need to worry about the actual processing of trades and protocol translations. These are complex parts of the process, but ones that hold no commercial advantage except in the need to move and translate messages quickly, accurately and reliably.

At the end of the day, firms need to remember that it might not be as challenging, or you might have to give up some control. BUT buying allows firms to concentrate on where they can add value and differentiate themselves.

Maria Netley

Business Development Manager at RA and Co Chair of FIX Trading Community (EMEA)