The Economic Times reported last week that the Securities and Exchange Board of India (Sebi) plans to introduce a temporary trading limit of 8-9% for all stocks to address problems related to erroneous orders. Earlier this month, an erroneous order by a trader at Emkay Global Financial Services Ltd had caused havoc at the National Stock Exchange. Soon after, Sebi asked its risk management review committee to suggest changes to the existing market structure.
Back then, this column had pointed out that one way to minimize damage during such episodes is to narrow the price limits within which traders can place orders. See (tinyurl.com/nifty-flash-crash). According to a Bloomberg report, Indian exchanges, too, asked Sebi to narrow price limits. It now appears that Sebi is seriously considering this proposal. But, according to an expert on market structure who did not want to be identified, this solution is fraught with problems. Narrowing price limits will minimize losses from an erroneous order, but will also lead to unintended consequences.
Currently, for India’s most liquid stocks, while there aren’t any price limits, traders can place orders only in a price band of +/- 20% compared with the previous day’s close. This price band is flexible, and when the traded price approaches the +/-15% level, the band is extended. While there is no official statement from Sebi yet, narrower limits could mean that traders will not be able to place orders at a price that is beyond +/- 8-9% compared with the previous day’s close. Of course, this price band will remain flexible to allow price discovery. However, in the event of an erroneous order, losses will be curtailed at 8-9%.
But as the expert correctly points out, narrowing price limits works against the interest of irregular traders and will also lead to a further deterioration in the depth of the markets. Regular traders can easily cancel/modify orders and align them with a flexible price band. But retail investors often put in normal as well as stop loss orders at prices that are far away from prevailing levels, since they can’t monitor positions as often as regular traders. If these orders aren’t accepted because of narrower limits, such investors will end up losing. More importantly, if fewer orders are permitted because of tighter price bands, the natural fallout will be that order books will get much thinner. As it is, the Emkay fiasco made it clear that Indian markets lack depth—transactions worth Rs.650 crore ($125 million) wiped out the entire order book of all Nifty stocks. Sebi should ensure that fixing the erroneous orders issue doesn’t result in other problems.
One way to do this is not tinker with the order book, but put limits on order execution. In other words, the order book can be populated with various bids and offers without any price restriction. But when a market order or an aggressive limit order threatens to disrupt the market by wiping away a majority of the order book, this order shouldn’t be allowed to be executed beyond a pre-defined limit. If, for instance, the limit is set at 8%, then this is how the Emkay situation would have played out. The broker had erroneously put in a market order worth nearly Rs.1,000 crore to sell the basket of Nifty stocks, which had caused most of these stocks to fall 20%. In a world with an 8% price limit on execution, trading in each of the Nifty stocks would halt after falling by 8%. Sebi must ensure here that any halt should happen across exchanges and even in derivatives segments.
This halt will allow exchanges to assess if the drop/rise in prices has been caused by an erroneous order or is news driven. Traders will get enough breathing time to cancel erroneous orders and orderly trading can resume quickly. But the problem with both of the above-mentioned proposals is that it will affect price discovery in cases where drop/rise in prices is because of a news development. Consider the sharp fall in shares of Satyam Computer Services Ltd after founder B. Ramalinga Raju in January 2009 confessed to misstating accounts. If trading had halted after an 8% drop, a number of traders and investors would have rued their inability to exit as quickly as possible. Because of this possibility as well as some other weaknesses of imposing limits, some experts such as Ajay Shah of the National Institute of Public Finance and Policy say that it’s best not to tinker with the existing system. Shah wrote on his blog that rather than prevent fat finger trades, it is better to realize that mistakes are an inevitable part of financial markets, and hence it’s important to focus on deeper initiatives that will make the market more resilient.
The market structure expert, too, says that it’s important to avoid draconian measures. But not doing anything to prevent Emkay-like episodes isn’t wise either. Sebi must take a leaf or two from the book of its US counterpart Securities & Exchange Commission, which approved a proposal from US exchanges to have a ‘limit up-limit down’ mechanism that prevents trades in individual exchange-listed stocks from occurring outside of a specified price band. While the mechanism is very similar to the solution of putting limits on order execution mentioned above, it includes many useful features, thanks to an exhaustive collaborative work by US exchanges, apart from the benefit of incorporating feedback from market participants.
For stocks above a price-specified threshold, there will be a 5% price band on execution. This band will be dynamic and will change depending on the average traded price in the past five minutes. If a stock suddenly rises/falls 5%, it will enter a ‘limit state’ for a 15 second period. During this period, trades can still happen, but only within the +/- 5% limit. If the rise/fall was caused by an erroneous order, it will give the trader adequate time to cancel the order, after which normal trading can resume. Trading can also resume if there is enough demand/supply to execute against the orders at the limit state. If, however, the order which caused the rise/fall is neither cancelled nor executed, then a five-minute trading pause will be declared, before normal price discovery can begin again. In a Satyam-like episode, no one would want to sell at a 5% lower price. In such a scenario, trading would halt for five minutes, after which it can resume. Like in the Emkay episode, trading can resume with a call auction.
The dynamic price limits as well as short trading pauses are useful features and have many merits. But even if Sebi doesn’t see much merit in them, the least it can do is to initiate a public discussion on this, to benefit from the feedback of market participants and experts.
Your comments are welcome at firstname.lastname@example.org