19 highlights
-
He lost his high-paying job and his bet on quick gains from the stock market went horribly wrong; his entire savings of Rs 25 lakh got wiped out on the new stock market fad of retail algorithmic trading, or retail algo.
-
The proponents of retail algo called them blatantly false, sponsored by the big boys (read: institutional investors); the more circumspect found merit in the concerns that had been flagged.
-
Clearly, the blooming market trend had retail investors divided. The ones with a slightly better understanding of retail algo or those who had made money during the bull run (still on, by the way) were all praise for it and the not-so-lucky ones were looking at it with suspicious eyes.
-
In a sense it democratizes the stock market for institutional and retail investors. The former, with their deep pockets, have for years been using algorithmic trading and co-location facilities (which allows them to place servers in exchange premises to reduce latency), putting them at a distinct advantage by enabling them to capitalize on the milliseconds saved in executing their giant orders.
-
The lack of regulations and the belief that retail algos can help one make money faster are at the crux of the problem. The growing popularity of algorithmic trading is such a draw that many investors, despite their limited understanding, view it as a casino opportunity. The regulator is behind the curve in putting in place even the most basic of checks and balances
-
Institutional investors in India have for over a decade and a half been getting in and out of stocks in fractions of a second in the hope of making profits. While on a micro-level these may be tiny profits, but the large positions they hold results in massive gains. So, even nanoseconds make a difference.
-
Retail investors have typically not been part of this high-speed game of getting in and out of stocks, but things have changed in the last four years.
-
The only thing that separates retail algo from institutional algo is co-location; the big boys can buy rack space on stock exchange premises to reduce the lag in getting market data and executing orders.
-
According to the Securities and Exchange Board of India (SEBI) annual report for fiscal 2021, the number of non-algo trades in the cash segment has fallen to 17.3% on NSE and to 30.8% on BSE. In 2019-2020, non-algo trades on NSE came in at 23.5% and at 37.6% on BSE .
-
This means for the cash segment, algo trades account for over 82% on NSE and close to 70% on BSE. This includes co-location, direct market access, algo and internet-based trading.
-
Thus, Indian markets are inching closer to how trades happen in the US, where close to 90% of trades are algo/ internet-based.
-
Fully automated algos need to be approved by exchanges while manual algos provide an application programming interface, or API, to users to write an algo which executes orders when prompted.
-
So, a majority of brokerage firms are just offering APIs for manual execution of orders. “They leave some measure of control in the hands of users and ensure that brokers have limited liability,”
-
Notwithstanding the rise in the number of retail investors using retail algo, SEBI does not feel there is any immediate need to step in for now.
-
This, despite instances in the past where algo and high frequency trading have led to manipulation of markets using techniques like quote stuffing (a practice of quickly entering and then withdrawing large orders in an attempt to flood the market with artificial demand). Such manipulation of the market by algorithms can lead to decreased liquidity, higher trading costs and short-term volatility.
-
This is not to say that manual trading is blemish-free. Like algo trading, it is also prone to the so-called fat finger, or the erroneous punching of an order, and mini flash crashes.
-
The more I spoke to industry participants the clearer it became that in the absence of standardization and regulations no two brokerages were approaching retail algo in the same way.
-
Some algos are being run as a computerized trading program that assures profits in certain situations. So in a way they are doling out advice.
-
Many investors are relying on back-testing, which goes back only as far as 2018 and is not an indicator of future performance. Many others are simply copying and adopting strategies of entities that are not under SEBI’s ambit.