Algorithmic and High Frequency Trading in India

Algorithmic trading or algo trading is the use of computer programs and software to execute trades based on pre – defined criteria and without any human intervention. High Frequency Trading (HFT) is a subset of algorithmic trading, and this type of trading involves buying and selling thousands of shares in fractions of seconds.

HFT came into spotlight about two years ago when Goldman Sachs sued one of their former employees for stealing code that was used in one of their programs used to execute this type of trade.

For a while, newspapers were all over this story, but with time the attention faded. This event helped bring to light something that most regular investors didn’t know about – that large institutional investors have been allowed to place their servers in the same building as the stock exchanges, and they are able to gleam at transactions a few milliseconds before the other market players do.

The NYT later did a full story on HFT and while the actual way HFT works remains a bit of a mystery, it did help clarify that these big players get trade information milliseconds before other players, and as a result they can trade on this information and trade on the spreads that exist at that point in time. That article had an example of how this affects share prices and they used a $1.4 million order on which they said the slow moving investors had to pay $7,800 extra because of high frequency traders which is about half a percent more on the trade than they would have otherwise had to. An ET article that appeared recently quoted NSE’s Chief Technology Officer stating that the impact cost of high frequency trading on the NSE is 0.07% – 0.08%.

Since these operations are rather opaque, and these are just two numbers I think it’s not reasonable to compare them, but in terms of the impact that they currently have on prices, I think a long term investor is not impacted a whole lot by the small price changes either in India or in the US.

However, in May 2010 – a flash crash took place in the Dow in which several companies and blue chips lost a lot of their value in a matter of minutes, and the NYT reported that shares of big companies like P&G and Accenture saw ridiculous prices like a penny or a $100,000. The prices were later restored to more usual levels.

The SEC investigated this, and said that while the starting point of the flash crash was a mutual fund legitimately hedging its position by selling mini S&P Futures – a confluence of factors, primary among which was HFTs getting in on the trades, and then suddenly stopping altogether caused this market distortion.

Although, there weren’t any reports of regular investors losing money because of these events, it does show that these computer trading programs pose risk, and can cause damage because of the way they are structured. Even in India – BSE canceled all the futures traded on muhurath trading this year, and at least an initial report blamed an algo trader from Delhi for causing havoc because of their trades.

Since no one wants to talk about actual volumes that can be attributed to algorithmic trading – it’s hard to say how big it is in either India or in the US, but I think the fact that no one wants to disclose it shows that it must be reasonably large.

These were the similarities between the US and India, but there is a difference too. It appears that while it’s only a question of how profitable these operations are in the US and whether the profits of firms that engage in this runs in billions or tens of billions; in India – the profitability is itself questionable.

I don’t know how far this is true, and I’m certainly skeptical about it – but maybe they don’t have enough volumes or enough of an edge to be as profitable as they are in the US.

To summarize what we know:

Algo and HFT exists in good volume in India, and while the difference it makes on an individual trade is very low, it is capable of causing wildly large market movements. So far, we have not seen any real damage because of these swings, and exchanges seem to have things under control.

Now, as far as a small investor is concerned, unfortunately even if you don’t agree with the argument that HFT provide liquidity to the market, and are an essential part of the market there is not much you can do about it and you need to really focus on stuff that you do have control over.

What you can do is always place limit orders, and not sell in this kind of panic – that’s all which is in your control, and rather than fretting over half a percent I think small investors are better served focusing on things that are in their control.

Personally, I don’t see any utility in HFT and find it unfair that someone should get information before others, if only for a few milliseconds, and don’t buy the argument that they really do contribute to liquidity. The markets were working quite fine without them for decades, and that someone by virtue of their size gets an edge on other market participants seems unfair to me.

This post was from the Suggest a Topic page.

29 thoughts on “Algorithmic and High Frequency Trading in India”

  1. Obviously this author has no idea of what HFT is. This is analogous to someone in 1900 saying that he sees no need for cars and horse drawn carts can do the job well enough.

  2. Algo trading is a threat for small / retail investors. If markets r regulsted ny bodies such as SEC,SEBI then why is there no policy in place for algo trades? It appeard to be an intentional ovrrlook by regulators who cinjure wd big investors/ traders for their own interest in the blind eye of law.

  3. I was looking for topics on algo/HFT in India when I stumbled on this post. I am sorry to see that like most of our Indian bloggers/writers the article is full of anecdotal evidences from other news sources. HFT trading requires a deep understanding of market microstructure which is clearly lacking.
    The kicker is: “What you can do is always place limit orders” . If the writer had understood HFT he wouldn’t advice doing this.

    1. I fully agree with you about anecdotal evidence and me not having any insight on HFT.

      But what’s wrong with placing limit orders? How is that harmful to small investors?


      1. “Always place a limit order” is just gross generalisation of things. A complete answer will be a bit long.
        There are broadly 3 types HFT:
        1. Market Maker
        2. News Trader
        3. Algo technical and corelation based trader

        The whole providing liquidity arguement is being pinned to type 1.

        Crashes in themselves are being caused by type 1 & type 3. Type 1 when withdraw their bids/offers or rapidly hit bid/offer cause a violent move. Type 3 being the machine it is takes the movement and tries to do arb/ Momentum trading thereby excarabting the violence. And you get a full fledged crash.

        Under normal conditions, order type – market/limit wont make much difference. Even after all the hoopla over the loss caused by HFT prices are good. For small investors the price difference will be very minute when the market making HFT tries to close the bid-offer gap. Its the big traders who are hurt when their order is front run – thereby they end up paying more for a trade.

        Under exceptional conditions, market/limit both are a bad idea. Market orders for obvious reasons. Limit orders cause they might get hit and small investor (stress on small) migh end up getting margin called – even when they dint do anything wrong. You can check out a recent crash due to Knight Capital market making HFT. You wont find this in any mainstream article but they seem to be hitting the bids/offers with 100 orders each. So irrespective of order type, once you were in there was no helping.

        This beings me to the over-generalisation part. It is impossible to predict when the conditions will turn exceptional so worrying over market/limit order is just inconsequential. But if market has gone ballistic even limits wont help.

        1. Hi Vicky,
          How can a small to medium day trader use the algo trading for his/her advantage..i mean can we buy the software ? whats the reliable company that sells the software? thanks…

          1. Kris, there are certain softwares which exploit algo behavior. They employ market microstructure to defeat the algos at their own game by analyzing order stream in the market. Though none of them are available at the small/medium level. If you have a good understanding of the microstructure fundamentals you can to a certain level exploit them without need for softwares.
            I forgot to make one point in the last post. Putting a limit orders are also dangerous if we were to talk about algo behavior. They are made acutely aware of people’s tendency to place limits at certain points. And once limits are in the books everyone can see your intentions.

  4. We have developed business strategy of using simple technique of trailing of stop loss and target to the advantage of traders and corporate. It is neither HFT nor full algo kind of trading discipline. Considering the plight of loosing investors it makes the situation easy to operate the trade. In the simulated model it has achieved 7 trade successful out of every 10 trades. It has nothing to do with trend etc just works on arithmetic of the business. You can set the algo formulas without even writing a single line of coding. But simple English like sentences will help you to set your trading discipline for the day trading. Interesting is it n’t.

    1. Rajbabu :- I am interested in Algo trading as I don’t want to sit entire day before the terminal. Please tell me hoe can I do?

  5. Great article..can we also have an article on the growing clout of Non-HFT strategies in India. Again a subset of Algorithmic Trading, but here latency is expected to be in milliseconds not Micro seconds. Non-HFT Algo strategies basically consist of Technical Analysis/Fundamentals/Quants i.e some intelligence put in the strategies rather than just blasting of orders. Many of the brokers are going for automation of technical analysis due to one simple fact: Arbitrage is proving to be too expensive and the returns are too low.

  6. Algorithmic trading in India is very much present. In US, about half of all trades in equity markets are accounted for by algorithmic trading.

    @Ashok: In India, all major exchanges offer co-location facilities, through which, traders can pay a fees and setup their servers in the same building as the exchange, and connect to the exchange using high-speed network. This allows the trader’s algorithms to execute trades within a few milliseconds. This isn’t a US only phenomenon.

  7. I think soon this will come in the radar of SEBI. You are right that till now, it has not affected the small investors. But this may change in the coming days. So it makes sense for the market regulator to critically take a stand on it.

  8. Insighful post Manshu ! You are entering the fringes of Forbidden territory !!! I am sure that it would be a nice story for a movie !!!!

  9. While I agree with your comment about focusing on the things we can control I also think that this type of trading should be illegal.

    What benefit does this serve investors? If it only serves to make the banks and trading institutions money one has to question the ethical ramifications. I am not a supporter of the OWS protest but it is information like this that is giving them a reason to protest.

    1. Algo/HF Trading is neither designed nor are beneficial to investors.

      Investors should consider instruments/strategies like Futures & Options and Algo/HFT, as fire.

      As per their financial plan, investors should stick to the real assets that they understand viz good equity shares/MF, debt, metals (ETFs, eSeries of National Spot Exchange), and real estate.

      1. A lot of people write in to ask about Futures and Options to me as well, but generally my recommendation is to stay away and not get into that type of speculation. But the interest is rising and I think I’m due for a post on them. Do you get a lot of inquiries on F&O also?

        1. Yes I do get queries on F&O.

          Most people who are interested for F&O are those who want quick bucks.
          These “investors” (or traders!) make little profit here & there but end up incurring a huge loss then never see the market again – missing the huge profits that equity market offers them over long run.

          I feel that F&O should only be used as a hedge for equity holding, which was the primary purpose that F&O segment was created for.

          I welcome your post of this topic as you are known for making complex concepts simple to understand πŸ™‚

          1. Yes, I can totally relate to that and I see that happen to friends and relatives all around me. Make a percent here, percent there and then lose – in fact one of my fav fin profs in college used to say that a long term investor is a trader whose stock has gone down in value πŸ™‚

            Also, people don’t correctly judge the impact of leverage or margin that options bring in and esp. while trading in futures I feel that they don’t realize how quickly the market can turn and make them book heavy losses.

Leave a Reply

Your email address will not be published. Required fields are marked *