High-speed trading, low-speed start

Algorithmic trading has a long way to go in India

The expert committee on ‘Mumbai as an International Financial Centre’ recommended that the Indian markets open up to algorithmic trading and direct market access by end-2007. Four years on, though Sebi has set up the framework and exchanges have put in place the systems, volumes through these segments have not really picked up.

Algo trading or automated trading is based on technology-driven pre-programmed mathematical model-based stock trading, which is quite popular in developed nations.

Some of the commonly used algorithms in the market place are: arrival price, time-weighted average price (TWAP) and volume-weighted average price (VWAP). Arrival price is the midpoint of the bid-offer spread at order-receipt time, and it also notes the speed of execution.

VWAP is calculated by adding the dollars traded for every transaction in terms of price and multiplying it by shares traded, and then dividing that by the total shares traded for the day.

MOC measures the last price obtained by a trader at the end of the day, against the last price reported by the exchange.

Implementation shortfall is a model that weighs the urgency of executing a trade against the risk of moving the stock.

Most algorithms of these algos allow customers to change the timing of execution, the rate of order-filling attempts at the beginning or end of the trading day, and the tolerance for the slippage of a stock from certain benchmarks.

Sandeep Singhal, co-head, institutional equities, EMKay global, said the new tools are used primarily by brokers for their proprietary trading and foreign institutional investors.

“Transactions like buying and selling of options by brokers for their prop books and simple time-weighted and volume-weighted execution techniques for institutions have moved to algorithms. Trading strategies themselves have not yet moved. Therefore, volumes in algo are nowhere near what they are in developed markets like Hong Kong and Japan.”

Developed market exchanges like the London Stock Exchange (LSE) had up to 40 per cent of the trades originating from alogorithmic trading systems as early as 2006-07. “It is estimated that around 40 per cent of the trades made on the LSE now originate from algorithmic trading systems. About 15 to 20 per cent of buy-side firms have adopted algorithmic trading broadly, and they are using it within the confines of their Order Management System (OMS) workflow,” a 2006-07 report by Tata Consultancy Services on algorithmic trading systems had said.

Indian exchanges have a lot of catching up to do. But they seem to be in no hurry. NSE.IT, a wholly-owned subsidiary of National Stock Exchange, says on its website, “Globally, algo trading is pegged at 3 per cent of the total turnover; we reckon similar size of Algo trading in India too, in 2-3 years.”

NSE.IT offers trading software like Algo Studio, Strategy Studio and FFFIX Engine for stock brokers that will help them in formulating a trading strategy and executing it. “The products used are ‘exchange neutral’, keeping in mind the convenience of stock brokers,” the website says.

However, the BSE, which is trying to make inroads into the NSE-dominated equity derivatives market, has accused its rival of deliberately slowing things.

Last year, BSE’s James Shapiro had alleged in a public forum that NSE is deliberately not clearing algorithms which it thought could divert volumes to BSE. However, NSE had argued that it was only putting in safeguards to avoid systemic issues.

Similar skirmishes have surfaced every now and then between the rival exchanges during implementation of technology advancements like smart-order routing, co-location servers, etc, leading to intervention by market regulator Sebi.

“BSE is desperately looking to win market share. NSE is equally desperate to protect it. We are caught in the crossfire,” says a broker who did not want to be identified. With uncertainty reigning supreme, he said many brokers are not yet building capacities.

“Recently, a domestic institutional broker approached a large foreign institution for some orders. Officials at the buy side player told the broker that his platform can’t handle even a fraction of our volume. And, this guy is supposed to be one of the larger brokers. That tells you the story of algo trading in India,” the broker added.

Some 15 foreign brokers and 10 top local brokers offer these services to investors today, according to industry estimates.

The lacklustre market over the past couple of years is also a bug in the algo, say traders. Volumes have been dropping steadily over the past few months. This has hit the efficacy of algo trades. Unless there is sufficient market depth, executing high-frequency trades may not be possible.

Ajay Pandey, vice president-institutional sales, Link Intime Securities Ltd, said, “Lack of depth in the market has reduced the frequency of usage of these algorithms. High-frequency orders require a certain amount of volume to be executed smoothly. At present this is not available.”

Pandey also points out that there is not enough trained manpower to handle such trades, especially in middle-level brokerages, restricting wider usage.