Sebi’s showcause notice to NSE in algo-trading case points to another shoddy investigation4 min read . Updated: 22 Jun 2017, 12:03 PM IST
What's worse still is that Sebi may have found evidence of undue gains by some stockbrokers in the algo-trading case but hasn't brought it up to help NSE save face
The Securities and Exchange Board of India’s (Sebi’s) show cause notice to the National Stock Exchange of India Ltd (NSE) generates a lot of noise and smoke, but lacks substance.
Considering that over two years have passed since it received an anonymous complaint about the exchange’s algorithmic trading policies and practices, the investigation falls woefully short of standards expected from the regulator.
Sebi’s main charge is that NSE did not provide fair and equitable access to the brokers using its co-location facility, and that certain brokers unduly benefitted as a result.
But strangely, Sebi’s investigation department hasn’t bothered to find out the financial gain the said brokers made by getting faster access to the exchange’s data and systems, or whether they made any gains at all.
Much of the other evidence provided is circumstantial, as far as the charge about undue benefits go. The fact that the regulator has failed to mention anything concrete on this leaves us with some disturbing inferences: the regulator is incompetent to analyse the relevant data; or that it has studied the data and hasn’t found evidence of undue financial gains; or, worse still, that it studied the data and found evidence of undue gains, but hasn’t brought it up to help NSE save face.
In recent years, Sebi’s track record with investigations has left a lot to be desired.
With NSE, an immensely important case, it had a chance to redeem itself. Besides, it had the luxury of time, the services of a forensic auditor (even though it was one appointed by the exchange at the instruction of the regulator), as well as a study commissioned by its technical advisory committee (TAC).
Despite all of these resources, the notice to the exchange only makes a passing reference to the possibility of undue gains. The reference is to a report commissioned by the TAC, which states that OPG Securities, a trading member, gained materially by exploiting NSE’s systems.
The committee, in turn, relied on the following observation to conclude that the member benefitted materially: It found that OPG’s success in getting its orders executed reduced dramatically when NSE changed its trading architecture from the TCP/IP model of communication to IP Multicast. While the former (transmission control protocol/internet protocol) scores high on reliability, it also results in minor differences in the speed at which market data reaches each trading member. OPG is said to have exploited this feature to be ahead of other brokers while receiving data from the exchange and using it to be the first to gain from arbitrage opportunities.
In mid-2014, NSE introduced the IP Multicast mode of communication, where data is sent simultaneously to many recipients together.
The report commissioned by the TAC and the show cause notice point that this significantly altered OPG’s activity on the exchange.
But all of this is vague.
Did OPG’s turnover and profit reduce substantially post 2014? Did its revenue and profit jump significantly in the period it is said to have taken advantage of the TCP/IP architecture? Was it an outlier in terms of gains made by brokers that employed similar trading strategies?
Finding answers to these questions isn’t rocket science; exchanges have all the necessary trade data and all the regulator needs to do is ask for it.
Of course, all of this is not to say that NSE is not at fault. Even if it turns out that OPG didn’t make material undue gains, the fact remains that there were shortcomings in the exchange’s policies and practices, and worse still, there was a concerted attempt to cover them up.
Besides, the forensic audit by Deloitte Touche Tohmatsu India Llp points to instances where emails and records of key staff were unavailable.
This includes records of two chief technology officers who served at the time of the alleged infractions, as well as a key employee who is alleged, according to some market participants, to have colluded with OPG. Sebi’s notice also claims that the exchange withheld information from Sebi and failed to co-operate with the regulator and other investigators. Undoubtedly, the exchange must face the music for these blunders.
But having said that, a key piece of information the regulator needs before it imposes a penalty on the exchange is the quantum of undue gains allegedly made by the brokers mentioned in Sebi’s show cause notice.
Besides, news reports suggest the exchange is keen to settle the case once and for all through the consent mechanism. But again, how will the penalty under the consent mechanism be determined unless the regulator has a handle on the size of unlawful gains?
Of course, not all is lost. Sebi can still pull up its socks and do the needful.
In a briefing with reporters on Wednesday, the new Sebi chairman, Ajay Tyagi, said the regulator will engage a forensic auditor to ascertain whether brokers made unfair gains. It’s difficult to understand what kept the regulator from doing so earlier.
But as they say, it’s better late than never.
In sum, the show cause notice is weak as far as incriminating evidence goes, even though it’s true that the circumstantial evidence points to lapses on the exchange’s part. The regulator should remedy this before it issues its final order.