A surge in algorithmic trading helped drive a significant increase in the share of US equity trading volumes executed via electronic platforms last year, according to the results of Gre
A surge in algorithmic trading helped drive a significant increase in the share of US equity trading volumes executed via electronic platforms last year, according to the results of Greenwich Associates’ 2009 US Equity Investors Study.
"After a chaotic period in which algorithmic trading strategies performed poorly and lost institutional equity trading volume to other methods of execution, algo trading came back strongly in the 12 month period covered in our research," says Greenwich Associates consultant Jay Bennett.
Overall, the proportion of US equity trading volume executed electronically increased to 36 per cent in 2008-2009 from 32 per cent in 2007-2008 – a shift attributable in large part to the pick-up in algorithmic trading.
More than three-quarters of all US institutions and 95 per cent of the largest and most active institutional traders use algorithmic trading strategies, which currently account for about 18 per cent of overall US equity trading volume.
Institutions that use algorithmic trades employ these strategies for 23 per cent of domestic trading volumes, up sharply from 17 per cent in 2007-2008. These institutions expect algorithmic trading to grow to 27 per cent of their trading volume by 2012, with banks predicting that they will be executing 33 per cent of their trading volume through these strategies by that time.
At least some of the renewed growth in algorithmic trading seems to be coming at the expense of direct-market-access smart order-routing trades (DMA). The proportion of institutions using DMA trades declined slightly to 58 per cent in 2009, and the share of total trading volume executed through these trades declined to 13 per cent in 2008-2009 from 16 per cent the prior year.
"Most algorithms lacked a pattern of historic data that could accommodate the unprecedented levels of volatility experienced in late 2007 and 2008," says Greenwich Associates consultant John Colon. "But many algorithms have now been redesigned to take the new data patterns into account, and institutions are once again embracing them."
Institutions appear to be moving slower when it comes to dark pools and crossing networks. Although the proliferation of dark pools has been the subject of considerable attention over the past several years, dark pools and crossing networks captured a flat 13 per cent of trading volume in 2007-2008 and 2008-2009, and the proportion of institutions using these systems remained steady at 73 per cent.
US institutions are using electronic trading platforms as a primary tool for lowering equity trading costs in a challenging market environment. The average rate paid by institutions on DMA electronic trades dropped to 1.6 cents per share in 2009, from 1.7 cents in 2008 and 1.8 cents in 2007.
The market’s most active traders – those that generate more than USD50m in annual trading volume – are paying an average rate of 1.5 cents on DMA trades. At the opposite extreme, institutions that generate less than USD5m pay an average of 2.0 cents. Banks pay the highest average rate of 2.6 cents per share while hedge funds pay just 1.3 cents on average.
Bennett adds: "With the average rate on a traditional ‘high touch trade’ holding steady at about four cents per share, institutions have been able to use electronic trading platforms to drive down their average ‘all-in’ blended commission rates and substantially reduce equity trading costs overall."
