AN ENRICHING EXCHANGE.

In a shrinking world, trading securities across the globe is now commonplace, with brokers routinely accessing trading venues thousands of miles apart. Olga Pokhvalova, head of sales & marketing of ARQA Technologies explains that innovative technologies are also more accessible than ever before – as long as you are prepared to look.

Located about 5,200 km from London, 9,200 km from New York and 4,800 km from Tokyo is the Siberian city of Novosibirsk. Not perhaps the first place you’d look for cutting edge technology solutions, but local firm ARQA Technologies has perhaps engineered another small step towards closer integration of traders and markets across the globe by developing a new trading interface with the London Stock Exchange (LSE).

This solution works in both directions: it obtains and supplies market data to clients as well as routing orders to the exchange through the same interface. In addition to the first iteration, which was based on the FIX Protocol, ARQA Technologies has implemented another version of interface which is called FIX2LSE. This was built on the basis of the exchange’s own API and was recently certified. This solution is the fastest in a range of similar solutions developed by the company. FIX2LSE will be used for the direct connection of external software suites or broker platforms. The interface also includes a fast pre-trade control module.

Past Experience

Along with new interfaces, ARQA Technologies draws upon technological approaches that have been developed traditionally for its domestic Russian clients. This experience of an alternative developmental history presents opportunities for users of the company’s trading platform (called QUIK) and also for any broker ready to try risk control approaches developed in Russia. When used by any trading platform the new interface will provide access to various risk management tools developed for QUIK, which in our opinion contain original ideas and address pressing challenges which have came to the fore recently – particularly issues of risk control.

Due to historic reasons the issues of risk mitigation and compliance have been the focus of Russian software developers for a long time. Risk control became a hot issue right at the time when financial markets started to develop in post-soviet Russia. From the mid 1990s brokers had to be extremely careful and consider all possible precautions against the multiple risks of the emerging Russian markets. They always had to make sure that clients had sufficient resources before placing orders and never let them trade beyond deposited cash or assets. Russian financial markets have been electronic from the very start. They have always been closely regulated and strict compliance was enforced. Such issues were immediately translated into standard requirements for trading software solutions.

Different models

There are four basic models widely used by Russian traders. One makes sure that trading is done within combined buying power limits. Another is used for margin trading and dynamic evaluation of collateral. For positions in various instruments spread over several markets and having diverse settlement dates within one account, a SPAN-like approach is employed. Finally, there is a model of risk control particularly suited for setting limits in proprietary trading. All of these models may be applied separately or in a combination.

Historically, solutions for pre-trade control have always sought a compromise between the complexity of evaluation approach (relevance and adequacy of checks) and the duration of transaction delay. Available solutions range from fairly simple checks for latency sensitive strategies to complex calculations embracing scenario evaluations for less low-latency dependent clients.

A breakthrough solution for comprehensive pre-trade control in situations when low-latency is critical employs a separate module which makes rather straightforward checks. This is sufficient due to the fact that all the more complex calculations have previously been done by the risk server and the result has been fed to the module. The decision made by the module is to let through, or block an order on the basis of previously processed data. By the moment a new order needs to be forwarded to a trading engine all essential parameters of the client’s position (including his buying power) have already been computed by the separate server. All previous changes have been accounted for and positions adjusted. This is really a pre-trade check based on post-trade data. This is an alternative approach to a more traditional concept of pre-trade risk control – using pre-set limits – which is fast but rather unsubtle. The fast pre-trade solution combines accurate assessment of the current position, including portfolio analysis performed post-trade, with a quick pre-trade check of the current order. This approach proved effective when pre-trade control was applied to low-latency infrastructures or used for HFT clients. Additional checks such as restricted securities’ lists or limits on volumes are often included as well. The check itself adds a fractional overhead (less than 10µs) to the time of order placement.

Far from perfection

All around us technological development is progressing so fast that sometimes one gets an impression that this is as good as it gets. But, frankly, it is far from perfect. There is still a lot of inefficiency. Here is a banal example. Almost everybody now uses an array of devices such as a cellphone, a smartphone, a tablet, a camera, a headset, etc. All these devices come with their own chargers, so one cannot help accumulating a lot of these things and is burdened with a heap of snakelike cables. Why can’t there be a universal charging device with a small set of plugs (or an even better solution) to connect to all of them? The need exists and will no doubt be resolved in the future, but not just yet. However, when the solution arrives, it may be something altogether unexpected, who knows?

For all its sophistication and complexity, the world of trading infrastructure and trading platforms may be somehow similar, though on a different scale. It is not inconceivable that in the future front-end platforms will be much more flexibly connected to trading engines across the globe. The existing universal protocol FIX already solves many connectivity issues. And from time to time here and there appear solutions which brokers and other financial market participants can use to their advantage. Technology knows no boundaries.