In a bid to shrug the underdog tag, BSE plans to slash trading time by as much as 97 per cent. The stock exchange is likely to ink a deal with a leading technology player this month to cut down trading time to just 300 microseconds from the current 10 milliseconds (one milliseconds is 1,000 micro seconds). This will help Asia’s oldest bourse to be among the fastest trading platforms in the world.
Currently, the London Stock Exchange (LSE) claims to be the fastest with a trading speed of 124 micro seconds. The trading speed of the National Stock Exchange (NSE) is in milli seconds.
The new technology, say experts, will improve BSE’s order handling capacity from 20,000 orders a second to 200,000 orders a second. The response time, too, may come down to 50-100 micro seconds, said a source with knowledge of the deal.
More From This Section
NSE, which runs one of the largest electronic speed highways connecting 2,000 towns and cities in India, is 10 times faster than BSE at its current latency. Latency is the time taken for order matching and confirmation of trade after a client keys it in. NSE handles 50,000 order messages a second for trading across asset classes with a capacity to scale it to 200,000 messages, retaining the round-trip latency in milliseconds.
However, technology experts believe exchanges in India are far behind their western counterparts. According to Hirander Misra, the chairman of UK-based Forum Trading Solutions and a consultant on automated trading, compare BSE's 10 milliseconds or NSE with LSE or the Nasdaq platform running on Singapore Stock Exchange, technology in India is far removed.
“Technology is an enabler. BSE has moved to offer co-location services, but they have not addressed the core issue of its BOLT (BSE's Online Trading) system. It is outdated, which is why BSE has had limited success in attracting foreigners to drive liquidity. In Europe, Chi-X was the first to gain tangible market share and challenge the monopolies, due to superior technology. Slower trading technology of Indian exchanges is a risk to market makers," said Misra. Chi-X Europe, a small trading platform run by Misra had challenged LSE to become the second largest equities trading venue in Europe in 2010.
Automated trading is picking up in India, but the problem is slower systems to disseminate strike updates. Slow reaction to information shows liquidity providers sitting on high risk and so they move away, say brokers. Around 20 percent of the daily trading volume in India comes from automated trading which is competitively low. About 55 percent of US equities trading volume now comes from firms using high-frequency strategies, says research firm Tabb Group.
While Indian markets have seen innovations like order-driven trading and central counterparty clearing, which once drove down costs, they are no longer subject to serious competition, believe experts. MCX-SX, the newly launched equity trading platform, is yet to offer co-location and high frequency trading in the segment, which will put its technology to test. Stock brokers say India has risk management, but exchanges will have to be reactive and their systems should be such that they alert them prior to the problem or predict it. Per-trade risk parameters will have to be robust. Global exchanges are moving towards this and the US is researching on predicting problems.
Correction |
MCX-SX has clarified that it offers algorithmic trading as well as co-location facility to its members, contrary to a report in Business Standard, which wrongly stated that the newly launched exchange was yet to offer these services. The errors are regretted. |
BSE AIMS TO DO A 'BOLT'
- To award core trading technology contract to Millennium Technologies or Deutsche Borse
- Technology upgrade will bring down BSE’s trading time from 10 milliseconds to 300 micro seconds
- To improve order handling capacity from 20,000 per second to 200,000
- To propel BSE to the league of fastest global exchanges
- LSE claims to be the fastest, with round trip latency of 124 micro seconds. NSE's is in milliseconds