Kathryn Zhao, Global Head of Electronic Trading, Cantor Fitzgerald
Electronic trading business is highly competitive. In my view, execution performance, speed and quality of customization, access to liquidity and system stability are differentiators of a competitive low-touch offering.
Over the last 20 years, the electronic trading technology landscape has experienced dramatic changes. Innovations in hardware, networking, and software have had an immense impact on the current state of art. To truly stand out from the rest of the competition, an electronic trading product must be performant from latency and throughput point of view and resilient to failure without compromising capabilities and quality of execution.
There is a prevalence of cutting-edge open-source libraries for almost everything you need to build. Tremendous amount of attention should be paid to reliability and failover. A comprehensive suite of automated testing and a sophisticated simulation environment is essential to validate and guarantee the quality of your product.
Buy-sides clearly demand better performance and predictability of execution results, without sacrificing the ability to source liquidity, transparency and control of their executions. Connecting to all sources of liquidities, selecting venues to route algorithmic child orders based on advanced analytics, regular review of venues selection, and transparency of order routing logic are all integral parts of best execution and drivers of quality execution performance.
Advanced quantitative models are the corner stone of a competitive electronic trading offering. It is the “brain”, which drives the algorithmic behavior. Taking into account both historical and real-time market data, limit order placement model determines when, where and how much quantity at any point in time throughout order lifecycle, venue ranking model makes informed stock-specific routing decisions, and volume profile model dynamically tilts towards either front-loaded or back-loaded.
You should choose the most suitable models for every situation you encounter. We treat our order placement problem as a Markov Decision Process solved by dynamic programming technique. It provides a mathematical framework that can naturally accommodate various sub-models to address different aspects of security trading. One of the most important aspects is the market microstructure, of which the spread dynamics, fill probability and adverse selection are all modeled with thorough statistical analysis. Short-term signals can also be easily incorporated into the framework.
At every state, an optimal decision among crossing spread, improving best quote, joining the best quote, and staying away from market will be selected in order to minimize trading cost. The usage of market microstructure model and dynamic programming technique makes a number of well-known optimal decision concepts become the natural outcome of LOPM in a quantitative manner:
1. Opportunistically cross spread when spread is tight or market is moving in the same direction as our trading
Explosion of fintech partnerships, data, analytics, cloud computing, and other digital technologies has dramatically enhanced private equity firms’ ability to streamline core business processes
2. Remove the order from the market if the market trades have significant adverse selection or market is moving in the opposite direction as our trading
3. Step into the spread to improve fill probability while still capturing spread
FIX connectivity/Vendor certification/ Client onboarding is one of the most crucial elements to get right in the electronic trading business.
Proper certification reduces number of issues down the road. Proper set up of FIX connectivity and tag translations creates a positive client experience, with reduced number of rejects and erroneous algorithmic behavior. In addition, expedited response to FIX inquiries enables electronic sales and coverage to respond to clients’ demands in a timely manner.
No two buy-sides’ needs are exactly the same. Modern, competitive electronic trading offerings are required to deliver customized solutions tailored to each client’s specific trading preferences.
With growing sophistication of quantitative models and ever-increasing electronic solution varieties, buy-sides face a unique challenge: how to choose a product that is straightforward to use and yet easy to customize.
The first question you should ask yourself when you evaluate a provider’s electronic offering is whether they have the capability to quickly customize their product to meet your requirements and what level of customization can they achieve. There are three types of customizations:
1. Customize within a particular algorithm and adjust parameters and behaviors based on price, time, or other market conditions;
2. Customize cross algorithms, i.e., switching between different algorithms based on certain signals or market conditions;
3. Bespoke algorithm offering based on buy-sides’ specifications.
All need to be done overnight, if not intra-day, to enable buy-sides to take advantage of market dynamics as they occur.
On top of that, you need to build a flexible experimentation framework that allows you to conduct A/B test and an advanced analytics framework that enables generation of performance statistics easily. Only by doing so you have a complete circle of capabilities for a modern customization framework.
Electronic trading is a “low-touch” business with “high-touch” service. The most competitive service model would be a combined approach of low-touch speed to market and high-touch level of consultancy. As the number of providers, algorithms, and venues grow, it is increasingly important for sell-sides to provide high level of customer service to their electronic trading clients. In order to deliver “high-touch” service, you need to have proper sales trading tools, capable FIX support function, and dedicated algorithmic support staff to answer buy-sides’ inquiries on the fly.
What is “high-touch” service?
• Stay on top of your trade blotter and watch your orders closely: orders that are not executing, orders that are falling behind the schedule, performance (slippage against Arrival, interval VWAP, etc.), orders that are causing impact, orders that are getting poor fill rates, stocks that are on the move, etc.
• Response to IB messages should be instantaneous
• Most importantly, catch things that clients care about (even if it is not good news) and catch it before clients catch it
• Provide actionable suggestions
An algorithmic product is only as good as the people behind it. “High-touch” service means your algorithmic coverage is clients’ eyes and ears, and provides them with value-added information in a frequency acceptable to them.
Enabling/papering client is step 1. Getting client to trade is step 2. Retaining and increasing clients business is step 3. Simply getting client to send an order on auto pilot mode is not something that will retain clients, and more importantly increase clients’ flow.