During the brutal market decline of Feb. 27 and the volatility that continued for the next four trading days, hedge funds relied on technology from their brokers, exchanges and service providers to execute orders. To what extent was their trading interrupted and how well did their service providers do?
"I think it was a wake up call for everybody, not only for the hedge funds, but for service providers," says Steven Harrison, president and COO at Imagine Software, a leading provider of a hosted real-time trading and risk-management service to about 150 hedge funds. "What do you think Dow Jones is doing right now and the New York Stock Exchange and Reuters and Bloomberg?" asks Harrison, who believes that all data providers are taking a closer look at the infrastructure and what capacity they can support."We didn't think volumes would spike like that overnight and so they did and so they can again," says Harrison. Because Imagine distributes data from vendors, it had issues too. "We had to make changes. We had to build out infrastructure. We had to respond to clients. It wasn't a pretty week," says Harrison.
With the stock market dropping 416 points on Feb. 27, following a sell-off in Chinese stocks and concerns about the U.S. economy, heavy equity trading volumes clogged the order delivery systems at the New York Stock Exchange, and taxed the market data systems at Dow Jones Indexes. When Dow Jones switched to a back-up system that recalculated the index, the Dow immediately dropped another 300 points in response as traders reacted to the true value.
"We saw a surge in automated trading particularly from our hedge fund clients when that refresh happened," recalls Greg Treacy, director of U.S. sales atNeonet, an institutional agency broker that provides direct market access (DMA) to active traders, hedge funds and other buy-side clients. There were firms on the buy-side that had bottlenecking because the strategies were telling them to send you so many messages in a compact amount of time, he says. "I think any bottlenecking was minute at that point, suggesting there were sub-second delays. While Neonet's technology held up quite well, says Treacy, he adds, "There were definitely market data constraints and capacity (strains), but nothing specific," he says.
According to Matt Simon, senior analyst at TABB Group, some of the sell-side execution management systems used by the buy-side shut down their pipes so they could catch up in the middle of the day. The higher volumes of market data slowed down the front-end systems, "so it would be logical to pull the plug on data feeds and focus instead on maintaining the trading piece of fulfilling executions," he speculates.
After reviewing FIX logs, Neonet's Treacy says there was decent capacity in the solutions to handle the surges and increases in volume and it looked like the bottleneck was with the NYSE. "It seemed like system wise, people were trading more, and that's why the New York took the hit," he says. As a result, execution reports were coming back late, says Treacy, noting that Neonet didn't receive execution back until 5:30 p.m. while he heard of brokers not getting back execution until 10 p.m.
On the other hand, Paladyne Systems, an ASP-based provider of integrated solutions for hedge funds, which partners with other market data, order routing providers, reports it was business as usual. "We didn't have any problems at all. Reuters was fine, NYFIX was fine. It was almost a non-event for us. Our infrastructure was fine. We didn't hear of any concerns," says Sameer Shalaby, CEO. 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 the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio