Even Outside Ultra-Low-Latency Trading, Technology Remains Key Competitive Factor

Although the entry costs for high-frequency trading can be prohibitive, and most market participants do not have the capital to write their own software, co-locate servers next to trading venues and invest in chip-based technology to run their algorithms, all trading firms—even those not sensitive to nanosecond-level latencies—must nevertheless continue to inject capital into their technology infrastructures, utilizing the full range of other technologies available to remain competitive, said panelists at last week’s Frankfurt Financial Information Summit.

“It depends on the direction of each customer as to whether a firm needs to invest, but generally speaking, you have to if you want to keep up with the market,” said Edward Strauss, head of sales and electronic trading at Baader Bank. According to Strauss, as Baader moves into equity sales trading and providing research on equity capital markets, the bank “currently needs a lot of technology and a lot of investment to be competitive.” He agreed with other panelists that a flexible order management system is one component of achieving a competitive edge, but argued that all components are key. “Other systems such as connectivity can be equally important. For example, if the FIX engine has problems, then all systems in the process will be affected,” he said.

Keith Wood, head of strategy and solutions at enterprise software giant Sybase, said trading firms must ensure consistent sophistication across all their systems—in particular, to ensure that risk, compliance and back-office systems can keep pace with the front office, and firms don’t find themselves saying, “Yes, I can do high-frequency trading, but can I do high-frequency settlement and high-frequency risk as well?”

Panelists highlighted cloud computing as an emerging future technology, but were unsure when it will reach its full potential. “Everyone is talking about [the cloud], but it all depends on how fast it can deal with data. I do think trading on the cloud will happen eventually, but it’s all about speed,” Strauss said.

Instead, Norman Hartmann, head of trading at Hamburg-based hedge fund Aquila Capital, pointed to new data types, such as sentiment-based, machine-readable news, that can be used to generate trade signals. “Given all the data sources there are, you can spend all day reading news. So it’s good to consolidate, and it can be an advantage,” Hartmann said.

Others agreed that news sentiment is an interesting avenue to explore, but were cautious about the reliability of sentiment-based measurement of some content from sources such as social media sites like Twitter, which don’t have the same authority as well-established news providers. “It differentiates trading strategies, which is good for the market, but I worry whether it makes sense because such tools do not identify who is tweeting,” Strauss added.