Hybrid model
Whether or not technology should be developed internally or bought from vendors largely depends on optimising human and financial resources, and on the specific applications of the technology. Trade execution algorithms need to be tested for different market scenarios, are proprietorial and might need to be recalibrated, so there is a strong case for building them in-house. Similarly, data gathering, analytics and interpretation can add alpha, so there are often commercial reasons for build rather than buy.
On the other hand, order management systems (OMS) are more generic and their basic requirement is that they are stable and function efficiently. It makes sense to install an OMS supplied by a vendor whose system has already been tested and which can be fine-tuned when new market conditions or regulations arise. A solid middle- and back-office architecture is a prerequisite for the successful application of complex algorithms designed to give a firm a competitive edge.
“There is an opportunity cost every time we decide to do something internally. You have to motivate the technology department, recognize there is an opportunity cost and aim instead to add more value in the areas that we can achieve alpha,” said a panellist.
There are, of course, many vendor products and services available to buy- and sell-side firms. The risk management process means they are more likely to choose the bigger suppliers. There is also an opportunity for niche providers, but clients need to be reassured that they can deliver and follow up. The best option is probably a hybrid: a selection of technology from a small number of vendors with domain expertise to complement the systems provided by a main supplier. However, it is important to ensure that their work and deadlines are synchronised. If not, a one-team policy is likely to be more efficient than a fragmented approach, as delays might offset the gains hoped for from a specific skillset.
Moreover, entrenched legacy systems can hinder the adoption of advanced systems that are more appropriate to meet the demands of new regulation and dynamic market conditions. They can resemble “calcified spaghetti”, almost impossible to unbundle and too expensive to replace despite their obsolescence.
Yet, firms must make a commitment to jettison legacy systems if they threaten to ossify their business activities. They will fail to reap the benefits of innovative technologies, which will undermine their commercial future.
Valuing Data
For instance, it is increasingly possible to harmonise multiple data sets that have been historically formed on the basis of a particular business or product. Visualisation and predictive techniques that can inform a targeted indication of interest, an optimal trading venue or a market making strategy could be integrated into a firm’s OMS and algorithmic systems.
Data collection and analysis is likely to be the main focus of research into new technologies. Regulators insist on the storage and publication of greater amounts of information in order to explain trade execution decisions. Meanwhile, the capability to tap into new sources of data and make sense of it will give buy- and sell-sides a competitive advantage.
Technology can be deployed to capture and analyse data more effectively, and present it in a visually appealing way. It is already good at managing structured data, but the greatest challenge now is to find ways to improve the collection and understanding of myriad unstructured data.
Artificial intelligence and machine learning will be an effective curating and analytical tool, but its adoption should not be a matter of faith, but rather of utility. Fintech innovations will also have important role. Brokerages and fund managers, as well as the established vendors, should retain the flexibility to adopt and promote them when appropriate and cost-effective.
We’d love to hear your feedback on this article. Please click here
