Via Business Line :

An average trade on the National Stock Exchange, the country’s premier stock exchange, was worth a little over Rs 30,000 in 2007-08. That is now down to Rs 20,000, a drop of around 30 per cent in a little over three-and-a-half years. The trade value has been declining consistently with the calendar year 2011 registering a 15 per cent fall compared with 2010. While some of this decline may be explained by a fall in stock prices, a section of the stock market attributes it also to the increased use of automated or ‘algorithmic' trading by investors**.**

The reduction in trade size started around the time the two exchanges facilitated high speed algorithmic trading. According to Mr Sandeep Singal, Co-Head, Institutional, Equities Emkay Global Financial Services, one reason for this fall in trade sizes is emergence of algorithm-based execution, wherein generally the system logic is to place smaller but more number of orders than fewer large orders.

Algorithmic trades are computer generated with little or no manual intervention. These programmes slice larger orders and send them in to the market in smaller sizes at the appropriate time so that the stock prices are not impacted. High Frequency Trading is a type of algorithmic trading where trades are executed at high speed to exploit price inefficiencies. Indian exchanges do not provide numbers on the volumes generated by algo-trading. But SEBI Chairman, Mr U. K. Sinha, pointed out on more than one occasion in 2011 that risk-control measures in algorithmic trading need a review.

This indicates that the share of computer generated trades in our market may be getting large enough to draw the regulator’s attention. Apart from shrinking trade sizes, there are other parameters that hint at the advent of automated trading. The number of shares in each transaction is down 10 per cent in the current fiscal compared with the previous fiscal. Investors were buying/selling an average of 120 shares in each trade in 2010-11, this came down to 107 shares this year.

On the other hand, the number of trades has doubled over the last five years. Trade sizes have declined even as retail participation in the market has actually waned. Data available globally show that trade sizes are reducing across markets, attributable mainly to computer generated trades. “Over time, the employment of simple algorithms for straightforward order-execution tasks became standard procedure, which is evidenced by the drastic decrease in the average trade sizes on major stock exchanges.

“On the NYSE, for instance, the average trade size is now only 200 shares, down from 1,600 shares 15 years ago; the average value of an order has fallen to $6,400 from $19,400 five years ago,” says a report by Deutsche Bank Research on the World Federation of Exchanges' Web site. Some market participants counter that other factors could be at work in trimming down trade sizes.

For one, the sharp decline in stock prices since market correction began in the last quarter of 2010 could also be reducing transaction values. NSE’s market capitalisation is down 30 per cent since last November. Increased market volatility that has made investors more nervous is also responsible for this trend, says Mr Monal Desai, Head Derivatives, Prabhudas Lilladher Group. He feels that though algorithmic trading has picked up momentum, it is still not a significant portion of the total traded volume. “But these trades have made markets more efficient than before since they iron out inefficiencies in volumes and prices,” Mr Desai adds.