The Securities and Exchange Board of India (SEBI), in consultation with the Finance Ministry, is expected to reissue fresh guidelines on high-frequency trading (HFT), popularly known as ‘algorithm trading’, after taking feedback from market participants, to effectuate which, will soon issue a consultation paper seeking market feedback.
The earlier proposal to tighten the algo trading rules, which had been put out by SEBI in August 2016, has been dropped by the ministry. The market’s contention was that the rules were framed without taking all aspects into consideration, which were not in line with global practices and did not have sufficient checks and balances, and also would have had an adverse impact on liquidity. To review the proposed norms, SEBI constituted in August, 2017 a committee on fair market conduct to suggest measures to improve surveillance of the market and strengthen the rules for algo trades.
SEBI’s new proposal would bring standardisation of the co-location (colo) facility. Most of the measures proposed earlier were regressive in nature. The new framework would be progressive in terms of equitable access without impacting liquidity. In 2015, SEBI began investigations after it received multiple complaints that some brokers had allegedly got preferential access to the National Stock Exchange’s (NSE’s) colo facility. According to the source, the consultation paper is being prepared keeping in mind the best global practices and considering cost benefit analysis, which was not effectively done in the previous proposal.
SEBI’s new framework would ensure that liquidity is not impacted. For that, the regulator may put a cap on “order to trade ratio” to give level playing field. This could be done by formalising market maker scheme. Under the revised proposal, the regulator is said to be focusing on a real-time surveillance system so that it could get minute-by-minute surveillance of algo trades and, thereby, detect and prevent malfunctioning. This could be done by improving the existing infrastructure and ramping up the systems by having an advance template of superior technology. Currently, there is no structured data available at a prescribed time interval to provide realtime feed for surveillance. Besides, SEBI also wants to have standarisation of colo facility, so that all participants would have fair and equitable access.
Publishing real-time colo-based latencies would help bring in transparency as well as provide benchmarking against global exchanges. The regulator is not in favour of allowing retail participation in HFT at present. It is not advisable as it would be highly risky for individual investors to use automated trading systems. The regulator wants to have guidelines in place first and it could perhaps consider domestic individual investors at a later stage.
SEBI also plans to address and mitigate the chance of flash crash and fat-finger trades. Algo trading is a software programme designed to execute automated trades on fulfilment of certain criteria. These are typically trading strategies that make use of complex mathematical models. The most common is arbitrage, which tries to profit from differential pricing of the same security at the same time on different exchanges. According to the SEBI data, a little over 80 per cent of the orders placed are generated by algorithms. Such orders contribute to about 40 per cent of the trades on exchanges (not all orders result in trades).
The review of the HFT regulation was triggered by the NSE co-location controversy, where some brokers and officials allegedly made illegal gains through preferential access to the server. After the controversy, SEBI had published a discussion paper and had proposed seven ways to level the playing field between HFT traders and others. These included revising the order sequence, introducing a minimum resting time between HFT orders, and uniform access to market data.
[Source: The Business Standard]
Sebi’s recent announcements on steps for strengthening algo trading through shared co-location has also boosted the sentiments of algo solutions providers and it is likely that this term will become a part of our daily lexicon sooner than we know.