HIGH FREQUENCY TRADING
The Information Arms Race
IN THE INFORMATION arms race speed is king. Traders and investment managers want ever quicker trading solutions; that is, customized solutions to address the requirements for ultra-low latency execution. The speed at which data can be processed and
exchanged between components on traders IT platforms directly affects the efficiency and overall speed of those systems. The phenomenon that limits this speed is known as latency. Minimizing latency – to a point where it is close to eradicated – is the design goal for IT specialists. But it is a design goal that is very challenging to
meet. Latency is introduced by every component of the platform, be it the microprocessors, memory and I/O buses that make up servers, the network interfaces and cabling that form local data fabrics, the fibre optics that connect geographically dispersed processing endpoints and by the processing involved in executing the trading application logic itself. The overall latency of an IT platform is the sum of many
contributors, including the vital messaging function or layer that binds the servers and networks (and the application logic) together. Trading firms continue to invest
significant sums in low-latency technologies, with connectivity and co-location a big focus (and a large proportion of IT budgets). Reducing propagation latency from the trading chain – by going with the fastest network provider and installing systems in co- location data centres – remains for many the simplest approach to staying ahead in the latency race (though in itself it’s not a universal panacea for business success). “We design each route
to measurably surpass the incumbent connectivity providers in all areas … speed, capacity, service … you name it. This is the nexus of Anova … our company thrives in this space, offering the largest trading houses the latency reductions necessary to maintain their top trading positions,” says Michael Persico, CEO and founder of Anova Technologies, a company exclusively concentrated on HFT and algorithmic trading clients. Nevertheless, trading firms are more carefully considering their investments in technologies, weighing- up their cost against the business benefits. There are those that want to operate at the cutting edge where they spend anything for speed and those where trading strategies can be optimised to deal with a certain level of latency, where technology investment is more measured.
The trend for co-location services has increased
significantly in recent years due to the growth of low latency trading internationally. Exchange operators, including NYSE Euronext, NASDAQ and most recently CME Group, have responded by opening their own data centres, thus assisting brokers and trading firms in meeting client demands. Regulators worldwide are grappling with the
HFT question. For example, Italy’s Borsa Italiana is introducing a fee structure that will charge participants more depending upon the number of orders they submit. This effectively introduces an HFT tax on traders who’s business model is predicated on a much higher order to fill ratio (so called “quote stuffing”). Yet, not so far away, the Swedish regulator claims that “the negative effects related to high-frequency and algorithmic trading are limited”. Companies like ICE are discouraging inefficient
and excessive messaging in their efforts to curb erratic high frequency trades that cause massive volatility in commodity markets without compromising market liquidity (targeting its most heavily traded futures and over-the- counter contracts). Accordingly, high frequency traders have made improvements to their algorithms and the quality of streamed orders in response. In the US, the SEC is looking into
techniques such as co-location, rebates that venues pay to spur transactions, direct market access where brokers let investors send orders to venues themselves, and whether the types of orders exchanges offer are being misused. SEC Chairman Mary Schapiro says that a large portion of equities
trading has little to do with “the fundamentals of the company that’s being traded.” It has more to do with “the minuscule aberrational price move” that computer- assisted traders with direct connections to the exchange can “jump on” in fractions of a second. “No one’s been able to prove whether high-frequency
trading is good or bad,” according to Larry Tabb, Chief Executive of Tabb Group LLC. “They [regulators] don’t have a consolidated audit trail, clean access to good data or the quantitative analysts to really analyse data, so they’re doing it through the resources they have – which is through enforcement.” For greater clarity, the CFTC is to form a subcommittee
tasked with defining and identifying HFT. Chief Economist Andrei Kirilenko will lead the panel. •
March 2012 79
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84