Regulating algos, road block or a detour

Sateesh Chandra

Sateesh Chandra

Share on facebook
Share on twitter
Share on linkedin
Share on whatsapp

Algorithmic trading: a phrase that has been ringing bells in the Indian equity markets for quite some time now. In lieu of its increasing accessibility to the retail investors, the Indian capital market regulator, SEBI, has put out a consultation paper that seeks to strengthen the regulation over algorithmic trading on stock exchanges. Before decoding this consultation paper any further let’s first understand the working and the history of algorithmic trading.


Alternatively referred to as algo trading, algorithmic trading involves the use of computer strategies to buy and sell stocks. It is also known as automated or programmed trading because pre-programmed computer strategies execute buy and sell trades based on parameters, instructions, market patterns and conditions. 


The goal of algo trading is to execute orders as quickly as possible. Humans take several seconds, and sometimes minutes to punch the buy and sell trades. Algorithms, on the other hand, execute orders based on predetermined market circumstances within milliseconds, before people even consider trading. Traders connect their pre-programmed algos to a broker’s trading terminal, which is connected to a stock exchange server. Pre-defined rules are typically based on timing, pricing, quantity, or any mathematical model. Apart from the potential for profit, algo-trading makes markets more liquid and trading more systematic by eliminating the impact of human emotions and errors on trading activities. Mobile trading is a type of algorithmic trading in which orders are placed through mobile apps. An advanced form of algo trading is order execution without human interaction.


Algo trading is not an out of the blue, 21st-century invention. Its former version, known as High-Frequency Trading (HFTs), was riding the wave of western equity markets during the 20th century. While many institutions took advantage of this new technology, Jim Simons, a hedge fund manager, has become sparklingly famous for achieving a staggering 66% CAGR over a period of 30 years. Jim Simons was the Chairman and CEO of Renaissance Technologies, a hedge fund which specialises in systematic and quant trading. Over a 30-year period from 1988 to 2018, Renaissance’s flagship Medallion fund, which is primarily managed for the fund employees, has the finest track record on Wall Street, returning more than 66% annually before fees and 39% after expenses. This record was mainly attributed to the high-frequency trades pulled off successfully based on complex statistical equations and mathematical programmes. 


One of the most attractive features of algo trading is its back-testing component, which allows individuals to run their strategy and see potential outcomes. You could also use simulation in algo trading to test your strategy in real-time without having to execute any actual trades. Also, risk management has become inherent via the means of the stop-loss feature. So the attributes like rapid execution, backtesting, eliminating human emotions and risk management attracted traders all over the world to leverage these methods in equity, derivative, and currency trading. Currently, 70-80% of the overall US equity markets account for algo trading.


However, the whole concept of algo trading isn’t pitch-perfect. Complete dependency on statistics and algorithms isn’t always supposed to turn out right. The following example of “May Flash Crash” explains the same.

By the year 2010, 56% of equity trades in the US were made by HFT. The May Flash Crash, which saw the Dow Jones drop 1000 points in a single trading day, was caused by a $4.1 billion computer-driven sale on May 6th, 2010. Within a 5-minute time span, the market lost nearly $1 trillion in value, as well as 600 points, before quickly recovering. According to the SEC and the Commodity Futures Trading Commission (CFTC), HFT businesses were substantially responsible for the meltdown. Hence, algos deployed without proper backtesting can only harm your trades.

With respect to its origin in India, SEBI introduced algorithmic trading in 2008 by granting institutional clients Direct Market Access (DMA). In summary, DMA enables brokers to supply their infrastructure to clients while also providing them with access to the exchange trading system, without the need for any involvement on their behalf. Initially, it was only available to institutional clients, not to retail investors. It is only in the last five years that retail traders have begun to use complex algos for trading.

Currently, in India,  algo trades as a percentage of the entire market is less compared to the western developed markets. Algo trading in the country is poised to further transform trading through the use of cutting-edge technology such as artificial intelligence and machine learning, and the utilisation of big data.

So why exactly does SEBI want to strengthen the regulations?

Well, the pandemic has led to the rise of retail participation through algorithms and a large proportion of retail investors subscribed to algo strategies through unregistered platforms (APIs). According to the extant regulations, exchanges are providing approval only for the algo submitted by the broker. However, for the algos deployed by retail investors using APIs, neither exchanges nor brokers are able to identify if the particular trade emanating from the API link is an algo or a non-algo trade. Unregulated/unapproved algos represent a risk to the market and can be misused for systematic market manipulation as well as to entice retail investors by promising greater returns. For retail investors, the potential loss in the event of a bad algo strategy is enormous. If there is no regulation intact to govern these unregistered algo trades, retail investors fall prey to daily high volumes of the trade. Moreover, since these third-party algo providers/vendors are unregulated, there is no investor grievance redressal mechanism in place.

To counteract the risk, the SEBI has proposed that all API orders be classed as algo orders and monitored by the brokerages. Furthermore, APIs that perform algo trading should have a unique algo ID, which will be provided by the exchange after it has been cleared through an approval procedure. Brokers must deploy relevant technological solutions to guarantee that sufficient checks and controls are in place to prevent illegal algo tinkering. Two-factor authentication has been proposed to gain access to algo trading via APIs.

Although this proposal is aimed at the greater good of retail investors, the third party API providers are raising concerns over their intellectual property being compromised. For approvals, third-party tech platforms and traders will need to share their algos and APIs with brokers. The concern is that some brokerages may copy the algo strategy. This has the potential to jeopardise the intellectual property of technological corporations. If the aforesaid proposals are implemented, a serious change in the trading volumes can be witnessed, since the scrutiny and approval of these pre-programmed strategies is a complex and time-consuming process.

With numerous amendments over the years, India now offers a solid chance for algorithmic trading due to several reasons including colocation facilities and advanced technology at both major exchanges; a smart order routing system; and well-established and liquid stock exchanges. Given the fast-expanding trend and demand for HFT and Algorithmic Trading in the economy, SEBI must increase financial literacy initiatives for investors. The importance of understanding market dynamics and fundamentals is not diminished by technological advancement. Aside from that, pitfalls await people who pursue technology for the sake of appearance of tech-savviness.