November 16, 2012

In this Q&A, Advanced Trading reached out to Tim Decker, senior product manager, Electronic Trading at San Francisco-based Advent Software. He offers insights into different technological crashes, and breaks down what the buy and sell side can do to mitigate the risks to their trading partners, as well as their own internal systems.

Advanced Trading: Are your buy-side clients concerned about mini-flash crashes occurring in stocks and ETFs? If so, what are their concerns?

Decker: Our clients have two types of 'crashes' on their minds due to recent events in 2012, as well as their institutional memory of the events in 2010. The first of their concerns are systemic crashes; these would be most like what occurred in 2010 where a futures trade quickly spread to cause significant price movement across the broader markets and crossed asset classes.

The second and possibly more top of mind are the technological/systematic crashes that have occurred for example with the Knight or NASDAQ (Facebook IPO) issues in 2012. These technological 'crashes' were much more limited in scope, yet have buy-side firms re-evaluating their trading technology risk exposure – evaluating both in-house systems and their trading partners approach to the automated trading environment.

What can buy-side trading desks do to protect their orders when they are using algorithms, software and services from third parties?

Decker: First and foremost, many of the traders I've spoken with are very aware that placing limits on their orders (or price limits in algorithmic systems) is a simple tool to help mitigate risk in the unfortunate event that the market swings wildly due to a 'flash-crash' of any type. The next thing we're seeing more of is 'kill-switches'. These systems give a firm the ability to immediately disconnect their trading systems from the outside world in the event of a rogue or malfunctioning internal system.

Many sell-side firms either have these or are implementing them. Certain types of buy-side firms could benefit from implementing a 'kill switch' system within their walls as well. Traders can also limit the amount of an order that is exposed to algorithmic strategy at any given moment. This will mitigate the risk that their whole order would be impacted given a malfunctioning strategy or system. Lastly is a strong system of human oversight whether real-time monitoring of the trading systems throughout the day or the use of TCA or broker provided reports to insure that strategy and executions systems are behaving as expected.

What steps can trading desks on the buy side take to mitigate exposure to technology-provider and system risk?

Decker: Of critical importance is knowing which systems within the walls of a firm can not only generate but also transmit orders to the sell-side. Once the audit is complete the next step is to understand the transmission and execution process for each product in the audit – Are there systems that auto-generate and auto-transmit orders? Do all my systems require human intervention before sending an order outside the firm?

This will provide a buy-side firm the ability to hone their own risk management programs to the systems that are subject to technological malfunction, and when necessary, work with their vendors or in-house teams to insure that these risks are mitigated. Once the in-house risks have been assessed and mitigated it is important to work closely with the desk's trading partners to understand what risk mitigation systems they have in-place (kill-switches, monitoring systems, etc.) as well as to truly understand how they manage technology risk at a cultural level (i.e., are the strong checks and balances in the technology creation, and deployment process).

How can technology help customers understand the different types of crashes that occur? For example, you mentioned that the May 2010 flash crash was very different than the Knight crash.

Decker: The largest challenge faced currently is that most technology solutions will only provide an after the fact assessment of the crash (and type of crash). Traders can gain intra-day insight with real-time pricing tools that alert them to abnormalities in the markets and real-time TCA tools that help them understand if the executions they are receiving are in-line with their pre-trade estimates.

These tools combined with some risk mitigation techniques when transmitting orders and a strong risk assessment program can help them respond rapidly to 'flash-crashes' and reduce their exposure to either the systematic or systemic event as soon as it occurs.

For example, an OMS can integrate some of these mitigation techniques, such as real-time pricing from industry leading providers and the ability to portion out orders to trading partners and algorithms, directly into the hub of the trading desk, thereby helping the traders and their firms be better prepared in the event of a 'flash-crash.'

Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in ...