Order in the matter of modification of client codes of trades in respect of NSEIL

IN THE MATTER OF MODIFICATION OF CLIENT CODES OF TRADES In respect of National Stock Exchange of India Limited Date of hearing: July 01, 2011 Appearances: For Noticee: Mr. J. Ravichandran, Director, NSE Mr. Ravi Varanasi, Senior Vice President, NSE Mr. Somasekhar Sundaresan, Advocate For SEBI: Mr. Rajeev Rastogi, Presenting Officer Ms. Anitha Anoop, Deputy Legal Adviser Mr. Navpreet Singh, Assistant General Manager Mr. Pradeep Kumar, Assistant Legal Adviser Order Under Section 12A of the Securities Contracts (Regulation) Act, 1956 1. Securities and Exchange Board of India (hereinafter referred to as ‘SEBI’) conducted an examination of modification of client codes done at the stock exchanges, pursuant to a letter from the Ministry of Finance (hereinafter referred to as ‘MoF’), Department of Economic Affairs, Government of India along with a reference from the Department of Revenue observing therein that substantial client code modifications had taken place in the derivative transactions at the National Stock Exchange (hereinafter referred to as ‘NSE’) during March, 2010.

2. In the light of the above reference, SEBI forwarded the above mentioned MoF letter to NSE on November 22, 2010 asking NSE to furnish details regarding the volumes of the transactions where modification of client codes were allowed by the stock exchanges along with its comments on the issues raised in the MoF letter. NSE submitted a reply to SEBI in response to this, vide letter dated December 03, 2010.

3. On an analysis of the data obtained from NSE, it was found that the volume of transactions wherein modifications were done, exceeded 55,000 crores in terms of value, during the month of March, 2010. It was also seen that certain brokers carried out a very large number of modifications in the month.

4. SEBI vide another letter dated December 29, 2010, advised NSE that modification of the client codes is allowed by the regulator to rectify inadvertent errors in punching the orders and hence such modifications are expected to be very few. In the letter written by SEBI, it was observed that modification of client codes of 500 trades a day by a broker, is not considered to be ‘in order’. Therefore, SEBI asked NSE to provide detailed information/ data regarding the client code modifications for separate segments such as cash, equity derivatives and currency derivatives. SEBI also wanted NSE to provide the details of its mechanism to verify the genuineness of modification of the client codes. Further, SEBI advised NSE to also conduct a special audit of the client code modification by the brokers who had carried out significant number of modifications in March, 2010 and submit a report to SEBI.

5. NSE replied to SEBI, vide its letter dated January 10, 2011. Pursuant to the examination of the details/ data submitted by NSE, SEBI issued a Show Cause Notice (hereinafter referred to as ‘SCN’) dated February 11, 2011 under Section 12A of the Securities Contracts (Regulation) Act, 1956 to NSE, wherein it was alleged that the NSE has failed to provide the information and clarifications sought by SEBI. Further, the SCN also alleged that NSE failed to conduct the special audit of the client code modification as advised by SEBI. The SCN also noted that NSE failed to put an appropriate mechanism for the supervision of the modification of client codes.

6. NSE replied to the SCN vide its letter dated March 11, 2011. Thereafter, an opportunity of personal hearing was granted to the NSE on June 13, 2011. NSE sought for an adjournment, vide its letter dated May 23, 2011 as its advocate was not available on the scheduled date. The request of NSE was granted and accordingly, the hearing was rescheduled to July 01, 2011. On this date, Mr. J. Ravichandran, Director, NSE, Mr. Ravi Varanasi, Senior Vice President, NSE and Mr. Somasekhar Sundaresan, Advocate appeared before me and made submissions on behalf of NSE. They also sought further time of 10 days to file their written submissions. I note that during the personal hearing, NSE was asked to provide the data of orders entered by the brokers on behalf of non-institutional clients in terms of value. The written submissions were filed by NSE, vide its letter dated July 12, 2011.

7. The submissions of NSE, made during the oral hearing as well as in its written replies dated March 11, 2011 and July 12, 2011, are summarized in brief below:

a. Modification of the client code is an integral and operative feature of the stock market system, which is consistent across all segments of the market. Client code modification is permitted by both SEBI and the stock exchanges so as to take care of the human error that occurs during the entry of client codes while placing orders. The errors in the client code entry could either be on account of the broker’s mistake or on account of the client’s mistake. The error at the client level could occur when the client is issuing transaction instructions on behalf of his family accounts or trading accounts of affiliated companies forming part of a group or an Foreign Institutional Investor (hereinafter referred to as ‘FII’) representing various sub-accounts or mutual fund representing various schemes. The client may communicate a code at the time of placing of the order and after the trade is executed, he may want to correct the client code from among those belonging to the same family/ corporate group/ FII/ mutual fund.

b. The regulatory framework has the impact of reducing the incidence of client code modification. The mechanism has been in place over a period of ten years and the same does not require any recording of reasons or prior approval from the stock exchange for modification of a client code that has been entered.

c. Modification of client codes is automatically recorded in the system and penalty framework is implemented where the incidence of modification of client codes is in excess of the specified daily tolerance limits. Significant percentage trade value client code modifications are considered as actionable and attract disciplinary action proceedings. The percentage modification by the brokers are scrutinized and examined vis-à-vis the total trading activity of such broker over a given period, besides imposing penalty on a graded basis during excessive modification, in such post facto analysis,. Thereafter, the views of the brokers are sought and an explanation is called for. During the period of January, 2010 - January, 2011, notices have been issued to 82 members, in addition to asking the brokers to review their operations and monitor client code modifications. Large client code modification has been one of the areas on which there has been focus during the inspections.

d. It is difficult to know the intent behind the modification or to require prior approval of the stock exchange with recorded reasons, for every modification especially considering the fact that millions of orders are entered every day. The reason provided by the client for changing the client code is accepted at face value and NSE does not have the power to investigate the same. Review of client code modifications is a process of scrutiny on a post facto basis as evident from the SEBI Circular dated February 06, 2003.

e. Members have represented that in view of the complexity and dynamic nature of the client code modification and the sheer volume of orders entered on the trading system by the members every minute, it is practically impossible for them to maintain records of the reasons for every client code modification.

f. The penalty system implemented by NSE has indeed resulted in a significant decline in the percentage of orders modified by a change in the client code. The percentage of orders modified has now come down to as low as 0.01% in terms of number of orders, on updation of the data by NSE since March 11, 2011, i.e., the date of their reply.

g. There has been no failure to comply with any advice given in the letter dated December 29, 2010. The response by NSE dated January 10, 2011 was a comprehensive reply.

h. Nothing contained in the SCN warrants issuance of any direction, whether under Section 12A of Securities Contracts (Regulations) Act, 1956 or otherwise. The framework governing modifications of the client code is identical across all stock exchanges. If SEBI felt that the penalty framework that was in vogue was inadequate, SEBI may modify the framework and increase the scale of penalties as also done vide its circular dated July 5, 2011.

8. I have carefully considered the SCN issued to the NSE, the replies and the written submissions made by NSE including those made during the personal hearing.

9. Considering the above discussion, the issues that arise for my consideration are: a. Whether NSE has failed to provide the information/ clarifications sought by SEBI? b. Whether NSE conducted a special audit of the client code modification as advised by SEBI? c. Whether NSE has failed to provide a mechanism to supervise modification of client codes? d. Whether the modification of client codes was allowed by NSE for reasons other than to rectify genuine errors? I now start dealing with the charges leveled as against NSE in seriatim:

10. Whether NSE has failed to provide the information/ clarifications on the issues as asked by SEBI? I note that SEBI asked NSE to provide the details of the mechanism used by NSE to verify whether the modification of the client code is done in genuine cases. Further, SEBI had also sought from NSE, the reason-wise break-up of the trades where the client codes have been modified by the top ten brokers during the month of March, 2010. SEBI had also asked various other clarifications from NSE, which are stated in the following paragraphs.

a. NSE in its reply has submitted that the client code modification was permitted by SEBI and the stock exchanges to take care of the human error in the entry of client codes while placing orders. NSE has submitted that it has introduced a mechanism of post-facto scrutiny with imposition of monetary penalties directly proportionate to the quantum of the incidence after considering the high incidence of client code revisions, in order to balance the interest of investors by providing the facility for client code modification and to ensure that the same is used only for correcting order entry mistakes as evident from the SEBI Circular dated February 06, 2003. NSE also submitted that, when the UCC (unique client code) requirement was introduced in 2001, the facility for modification of the client code immediately after closure of market hours by uploading a file in a specified format also was introduced. NSE in its reply has given a detailed picture of the mechanism of client code modification and submitted that the timeframe for revisions is narrow and presently there is no scope for modification of any client code entered on the exchange system after 16:15 hours on any trading day. NSE has also submitted that the modification of client codes is recorded automatically in the system and the penalty framework is implemented automatically. Further, the brokers are charged with appropriate penalty on monthly basis where the modification of client code is found to exceed the specified daily tolerance limits. The penalty structure is revised and updated consistently. NSE has also submitted that the incidents of modification of the client codes is analyzed on a regular basis and upon review, if the client code modification appears excessive, in terms of the traded value, disciplinary proceedings are initiated against the brokers who have exceeded the specified limits, in order to ensure that the facility is used for rectification of genuine mistakes during the order entries. Modification of codes in the case of large value trades by non-institutional clients are also a basis for initiating disciplinary action. Thus, the modification of client codes in large number has been one of the areas focused by NSE, during its inspections.

b. NSE has further submitted that the appropriate metric for assessing client code modification should be at the order level, as the error is committed while placing the order and that the same order may result in several trades. Further, while correcting the error, each trade that resulted from the order has to be modified.

c. The SCN has also alleged that NSE has failed to clarify various issues which are cited below: i. Whereas in the month of March, 2010, the client codes were modified in 1.58% of the institutional trades, the modification was not necessary in the remaining 98.42% of the trades. NSE was asked to give the reasons for the same. NSE has submitted in its reply that the percentage of institutional trades where client codes have been modified vis-à-vis total institutional trades was 1.58%. Further, the modification of client codes in respect of institutional trades is done by the trading members, for the allocation of the trades to various schemes of mutual funds or to sub-accounts of FII. If the fund manager of a mutual fund/ FII has already decided as to which sub-account or scheme the order should be placed under, before giving the orders to the trading member, then there is no need for further change in the client code. If he has not decided, he may ask the trading member to place the order in a single code and provide the client codes for the actual sub account/ scheme subsequently, which practice is in accordance with the SEBI Circular dated October 26, 2004.

ii. Why should a client suffer for an incorrect entry of the client code made by the broker? NSE has submitted that the reason for providing the facility of the client code modification was to ensure that the wrong client does not end up getting an obligation. It has also been submitted that the client is not asked to fulfill an obligation he was not meant to incur, irrespective of whether the error is made by the broker or the client, provided the client has pointed out his mistake within such period when the broker can carry out the modification of the client codes.

iii. Why client code modification is more endemic in large size trades? NSE has submitted that institutional transactions are of larger value when compared to the non institutional trade. Therefore, the institutional trades may have also contributed to the higher average size of the modified trades in comparison with the non-modified trades. This argument of NSE is not supported by the facts of the present matter wherein the percentage of institutional orders modified to the total order modified both in terms of volume and value, during March, 2010 was 9% and 12% respectively.

iv. How the modification of client codes becomes illegal, when the incidence of the activity is high? It is either legal or illegal. Or, it is legal in certain circumstances and illegal in certain other circumstances. NSE has explained that the modification of client codes does not become illegal when the incidences are higher in number. The transaction including those that involve the modified client code, continues to be valid and legal. Further, it is not possible that too large a percentage of such modification could be defended by the trading members as being caused due to errors in the client code inputs and that such facility has been provided only to correct the client codes that are erroneously entered. Therefore, modifications beyond a specified percentage attract penalty and disciplinary action, as per the Circular dated February 06, 2003, issued by SEBI wherein the exchanges were asked to provide for slab-based penalty with higher incidence attracting a higher penalty.

d. In view of the replies given by NSE, it does not appear that NSE was reluctant or slow in providing information/ clarifications asked by SEBI.

11. Whether NSE conducted a special audit of the client code modification as advised by SEBI?

a. The SCN also alleges that NSE has failed to conduct the special audit of the client code modification by the brokers who had carried out significant number of modification in March, 2010 and submit a report to SEBI along with an analysis of the data, insights into the problem, including the measures taken to reduce the occurrence. NSE in its reply has denied the allegation of not conducting review/ audit. It submitted that majority of the data was available with NSE itself and certain additional information was sought from the respective brokers. The analysis based on these data was submitted to SEBI by NSE vide its letter dated January 10, 2011. NSE has also submitted that it carried out independent verification of the contents that were submitted vide its letter dated January 10, 2011, through subsequent visits to the offices of the respective brokers. NSE has submitted that it has adopted a fair and reasonable approach in its analysis and communication thereof to SEBI. An analysis of the client code modifications done by the top ten brokers on the NSE was also submitted to SEBI, highlighting the details of the modifications done by each trading member, along with the reasons furnished by the members for such modifications.

b. Having considered the above, I note that although NSE has submitted an analysis of the data collected which it terms as audit, the same does not contain any supporting papers or requests of the clients for modification of client codes. Further, NSE has also not provided SEBI with a thorough analysis of the reasons/ factors contributing to the large numbers of client code modifications.

12. Whether NSE has failed to provide a mechanism to supervise the modification of client codes?

a. It has been alleged in the SCN that the client code modification has been done for other purposes and not to rectify genuine errors alone and NSE had not placed any mechanism to verify/ supervise such modification of client codes. I have seen the table of trades modified by top ten brokers during March, 2010 in the derivative segment of NSE, submitted by the noticee. The table has been extracted below for reference: Table A S. Broker No. of No. of trades No. of trades modified No trades modified of top within a pair of clients modified 10 clients with largest number of modifications 1. East India Securities Ltd. 38,661 29,073 6,713 2. Aadya Trading & Investment Pvt. Ltd. 30,109 14,061 1,447 3. Master Capital Services 18,708 8,718 714 4. Crosseas Capital Services Pvt. Ltd. 18,031 9,034 499 5. India Infoline Ltd. 16,406 14,486 3,697 6. Inventure Growth & Securities Ltd 15,285 9,416 1,387 7. Open Futures 13,777 11,793 2,346 8. Labdhi Finance Corporation 11,197 8,073 725 9. Maverick Share Brokers Pvt. Ltd. 10,498 4,168 495 10. Mbl & Company Ltd. 10,172 7,019 3,991 b. NSE has submitted in its reply that, it is not appropriate to look at absolute numbers. A broker who has a higher volume of trading is likely to have greater incidence of client code modification. Similarly, a broker with a higher institutional client base is likely to have client code modifications on account of institutional reasons. Further, the error at the client level could occur when the client is issuing transaction instructions on behalf of an FII representing various sub-accounts or mutual fund representing various schemes. The client may communicate a code at the time of placing of the order and after the trade is executed, he may want to correct the client code from among those belonging to the same family/ corporate group/ FII/ mutual fund.

NSE has further submitted that six of the brokers listed in table A above, do not have any material institutional clientele and therefore they did not have any institutional trading activity. It also said that as the value of trades where the client codes were modified was large as seen from the above table (the table has been extracted from the data given by NSE to SEBI and produced along with the SCN), disciplinary action proceedings have been initiated against the six brokers of the said total ten brokers.

c. An analysis of the data for trades relating to March, 2010 as provided by NSE, revealed that the modification of client codes was concentrated among a few clients and brokers. The summary of such analysis has been brought out below for reference:

- The top ten clients in terms of client code modifications accounted for 88% of total number of trades modified by broker India Infoline Limited. - In the case of four out of the ten brokers, more than 15% of the modification of client codes have taken place within a single pair of clients. - In the case of three out of the ten brokers, more than 75% of the trades were modified on account of only 10 clients. - Fifty one clients out of the said ten brokers modified 1,10,860 trades worth of 12,337 crore i.e. 72% of the total value of trades. On an average, each of the said fifty one clients had modified client code worth 242 crore in a single month approximately.

The above analysis can be further explained by taking an example of broker, East India Securities Limited. If one assumes that the month had 20 working days, East India Securities Limited modified client codes in 3,000 trades a day. Further, East India Securities Limited modified client codes of 6,713 trades between a pair of two clients. These details show that the modification of client code was concentrated among a few clients and brokers.

I note that five of the brokers referred in the table A above viz., Crosseas Capital Services Pvt. Limited, Labdhi Finance Corporation, MBL & Company Limited, Open Futures and Aadya Trading & Investment Pvt. Limited do not have institutional clients in the equity derivative segment. Further, three others, Maverick Share Brokers Pvt. Limited, Inventure Growth & Securities Limited and Master Capital Services Limited have not executed even a single trade on behalf of any institutional client. However, all these eight brokers, have modified large number of their trades. I have also observed that the total trades executed by the ten brokers as shown in table A above were 25,24,384 which constitutes 6.24% of the total trades executed in the month of March, 2010 at NSE. An aggregate of 1,82,844 trades were modified by these brokers, which means 27% of the modified trades during March, 2010. However, it cumulates only to 8.76% of the total value of trades at NSE, in terms of the value of the trades. I note that these brokers had accounted for about 31% of the total value of client code modifications. Therefore, I note that these figures do not support the argument of NSE that a broker who has a higher volume of trading is likely to have a greater incidence of client code modification. If the trend of modification of client codes was a normal phenomenon, then there should not have been such huge difference in the percentage points.

13. Whether the modification of client codes was allowed by NSE for reasons other than to rectify genuine errors?

a. I have perused the table produced by NSE, along with its written submission showing the modification of client code at the level of placing of orders. The table is extracted below for reference: Table B Month Total Total % of % of % of Total Total % of % of % of orders modif order Institut non-turnover in traded traded traded traded placed ied s ional institutio FAO value of value value of value of (lac) order modif orders nal segment ( orders of institutio non- s (lac) ied to in orders in Cr.) modified orders nal institutio order orders orders ( Cr.) modifie orders in nal s modifie modified d to orders orders in place d orders modified orders d placed modified Jan.,10 1654.52 2.13 0.13 15.02 84.98 1504673.45 35968.14 2.39 19.78 80.22 Feb.,10 1714.32 2.36 0.14 13.13 86.87 1386894.26 41817.48 3.02 15.73 84.27 Mar.,10 1246.28 3.41 0.27 9.03 90.97 1369154.96 55468.88 4.05 12.05 87.95 Apr.,10 1625.08 1.21 0.07 25.30 74.70 1399147.10 18019.87 1.29 34.08 65.92 May,10 2150.46 1.46 0.07 24.99 75.01 1687242.75 21309.92 1.26 37.31 62.69 Jun.,10 2032.95 1.74 0.09 24.09 75.91 1618865.34 22347.07 1.38 31.98 68.02 Jul.,10 1757.85 1.77 0.10 22.39 77.61 1448213.14 17329.39 1.20 28.80 71.20 Aug,10 1895.29 1.69 0.09 21.09 78.91 1615431.46 20543.51 1.27 30.84 69.16 Sep.,10 2247.04 1.78 0.08 18.97 81.03 1917057.41 25238.54 1.32 29.81 70.19 Oct.,10 2598.71 1.75 0.07 16.71 83.29 2061169.88 28175.80 1.37 25.56 74.44 Nov.,10 2757.11 1.73 0.06 17.30 82.70 1970606.80 25018.81 1.27 27.48 72.52 Dec.,10 2618.51 1.53 0.06 22.58 77.42 1544789.36 18491.28 1.20 32.28 67.72 Jan.,11 2522.8 1.31 0.05 20.78 79.22 1635704.64 15861.21 0.97 18.90 81.10 Feb.,11 2723.63 1.22 0.04 21.22 78.78 1657313.6 18101.14 1.09 27.59 72.41 Mar.,11 2704.5 0.86 0.03 25.34 74.66 1575706.62 12534.58 0.80 33.96 66.04 Apr.,11 2367.08 0.36 0.02 56.36 43.64 1305301.01 5112.84 0.39 75.51 24.49 May,11 2760.56 0.37 0.01 57.66 42.34 1324666.86 6743.32 0.51 83.75 16.25 Jun.,11 2651.09 0.39 0.01 57.70 42.30 1211744.75 5004.30 0.41 79.19 20.81

On a careful analysis of the above table, I note that the incidence of modification of client codes is substantially different across months. The total number of modified orders is showing an upward trend during the period January, 2010 to March, 2010. During March, 2010, the total modified orders as per NSE were 3.41 lakhs in number. However, there is a substantial decline by 64.52% in the number of the modified orders (i.e., 1.21 lakhs) in the very next month i.e., April, 2010 which is also the beginning of a new financial year. The total value of the modified orders has also shown a decreasing trend from 55,468.88 crore in March, 2010 to 18,019.87 crore in April, 2010, i.e., a decline of 67.51%.

I have seen similar variation in the data submitted along with the letter dated January 10, 2011 which shows the figures at trade level. I observe that the number of modified trades in the equity derivative segment increased from 3.85 lakhs in the month of January, 2010 to 6.80 lakh during March, 2010 i.e. an increase of 76.6% and the value of modified trades also increased from 35,968 crore in January, 2010 to 55,469 crore in March, 2010 (an increase of 54.2%). I note that the total trading volume and turnover in equity derivatives segment during the same period, decreased by 4.5% and 9% respectively. Similarly, the number of modified trades in currency derivative segment, has increased from 863 during January, 2010 to 19,395 in the month of March, 2010 showing an increase of 2,147% and the value of modified trades also increased from 461 crore in January, 2010 to 13,282 crore in March, 2010 i.e. an increase of 2,781%. I note that the total trading volume in the currency derivatives segment during the same period decreased by 19%, while the turnover has increased by 2.36%. The above analysis also shows that there is no fundamental difference in the date pattern when the metric employed is number of orders rather than number of trades’. This negates the argument given by NSE that the metric of number of orders should have been read.

b. NSE has submitted in its reply that it is not possible to discern with certainty any specific cause for the difference in the incidence of client code modifications across the segments and across the months of a year. NSE has also submitted that it has a rigorous and narrow time frame for the modification of the client codes, which approach is important to ensure speed, finality and accuracy for the client code entry.

I understand that modification of client codes is allowed to rectify inadvertent errors in punching the orders, as directed by SEBI vide Circular dated February 06, 2003, and hence such modifications are expected to be randomly distributed over months. Therefore, the number of modified orders should have remained more or less uniform across months with some deviations either to positive or negative side, under normal circumstances. However, the significant variation in the modification of client codes in the immediate months around March, 2010 as brought out in table B above gives rise to a genuine doubt whether the modifications have been carried out for purposes other than those to rectify genuine errors. I note that NSE is silent on the reasons for sudden increase in client code modifications during the month of March, 2010.

c. Before proceeding further, the relevant extract of SEBI Circular dated February 06, 2003, relating to the modification of client code is being reproduced hereunder for convenience:

“The stock exchanges shall not normally permit changes in the Client ID and would keep a strict vigil on cases of client code modification and would implement a monetary penalty structure that would escalate with the number of such incidences. Besides, the exchange may take necessary action against members making repeated changes. However, genuine mistakes may be allowed to be rectified.”

Thus, a careful reading of the above extract suggests that the change in client ID is generally not allowed except where there is some genuine mistake. The circular also mandates the stock exchanges to keep strict vigil on the instances of client code modification and implementation of a penalty structure. In the light of the reading and submission of NSE, I note that the practice adopted by NSE is ‘scrutiny of client code modifications on a post-facto basis’, with monetary penalties in direct proportion to the quantum of incidence in compliance with the SEBI circular. Regarding the requirement of not allowing the change in client code except for genuine mistake, I note that the term ‘genuine mistake’ has wide connotation. I have seen the submissions of NSE regarding the mechanism adopted by it, however, the same nowhere explains as to how the modification in client code is being carried out only in genuine cases. NSE has also submitted that it is not possible to know the intent behind these modifications, considering the fact that millions of orders are entered every day and the reason provided by the client for changing the client code has to be accepted at its face value. However, on perusal of the records, I note that no such reasons from the clients have been produced by NSE before SEBI, despite their being given sufficient opportunities.

Further, a careful study of the SEBI Circular would reveal that a two tier penalty structure has been provided for. A monetary penalty has been envisaged when there is an escalation in the number of client code modifications. However, in the case of repeated offences, further necessary action has been recommended. I note that NSE did not have any mechanism to penalize the repeat offenders.

d. I am of the view that the validity of the reason given while modifying the client code is required to be judged in the context of the factual matrix presented in each particular case. I have seen the circulars issued by NSE, copies of which were forwarded to SEBI. From the same, I note that NSE had a penalty structure for modification to client codes in the equity segment till February 17, 2011. The penalty structure was introduced in derivatives and currency derivatives segments by NSE from February 17, 2011. From the same, a conclusion can be drawn that NSE has introduced the penalty structure in its equity and currency derivatives segments, only after the issuance of the SCN dated February 11, 2011. In view of this, the submission of NSE that the modification of client codes is recorded automatically in the system and the penalty framework is implemented is devoid of any merits. I note that NSE, prior to the present proceedings has never expressed its difficulty in determining the genuineness of the reasons given for the modification of client code as required by the SEBI Circular dated February 06, 2003, to SEBI.

e. Now, let me deal with the argument of NSE that it is not possible to say what a genuine error is and what is not. It also says that it is legal to modify the client codes and then it does not matter as to the legality of the issue if number of such modification happens to be large. I am of the opinion that the number of errors is a strong indicator whether the errors are genuine or not. I note that the broker enters other fields as well viz. price and number of securities purchased/ sold. No modifications are allowed in for these fields even in case of a genuine error and the broker has to compensate the clients, if he commits a mistake. The cases where the broker had to compensate a client for mistake in data entry is truly rare. If we contrast this to thousand of errors committed in a month in the case of modification of client code, it is but obvious that there would have been some other reason than genuine mistake. Thus, NSE has taken a laid back attitude towards the problem and either totally ignored or perfunctorily imposed minor penalties to the brokers. It failed to apply its mind to the unusualness of the happenings. I therefore find that NSE acted negligently in discharge of its regulatory duties.

f. Under these circumstances, I note that the recurring instances of these client code modifications during March, 2010 are a serious issue for regulatory concern. NSE, being a self regulatory organization having the first level authority for supervision of the trading community within the securities market vis-a-vis the rules and regulations, should have taken the initiative to inform SEBI about these recurrent client code modifications as an aberration and further, should have devised a solution to curb the recurrence of these incidents. I note that the volume of instances of client code modification is considerably high and an appropriate mechanism to identify the repeat offenders and provide effective punishments should have been put in place by NSE. Monetary penalties are not effective if the gain to be made out of the offence is higher than the quantum of possible penalty and when there is no fear of any higher penalty on repeat offences.

14. I have considered the submission of NSE that its current regulatory framework of imposing penalties for innumerable revisions has the impact of reducing the incidence of client code modification and the structure is being consistently revised and updated. I am of the considered opinion that though NSE has taken certain steps for meeting the requirements of the SEBI circular, yet, it had failed in the first place to detect the increasing trend in the client code modification and failed to take any step towards repeat offenders.

15. NSE has submitted that a system of prior approval before carrying out modification would entail the need to exercise discretion in determining whether or not a particular request to modify should be allowed and that the discretion has to be exercised within a few hours from every single application for the approval of modification of the client code. It is also submitted that every trade is settled within two days (T+2 basis), and therefore by the end of the trading day, the obligations and rights of the respective clients of the respective trading member have to be crystallized and all such decisions would become a matter of statutory right. NSE has also explained that it would be impossible for any institution to provide a framework of prior approval for such a matter as it would be impossible to ascertain the veracity and accuracy of the reasons given by an applicant seeking approval for modification of a client code within such a tight time frame.

16. Having considered the facts and circumstances that are described in the previous paragraphs, I note that the total number of client code modification was definitely high during the month of March, 2010, compared to the other months. Therefore, NSE, being the first level regulator, should have noticed these aberrations through an independent analysis and should have brought this to the notice of SEBI. Such failure has occurred due to the complacency on NSE’s part that its systems and procedures are far superior. Considering the fact that it has a significant role in shaping the Indian stock market, NSE should have sensed the danger in the increasing instances of client code modification and should have adopted certain corrective measures in order to bring such incidents to a minimum. A Stock Exchange, being the first level regulator has to sense the aberrations in the market and alert SEBI so as to work out modalities for preventing any harm to the investors and to promote healthy development of the securities market. I note that there is a continuous decline in the modification of the client codes after March, 2010. The value of such modification is 0.86 lakh on March, 2011 and 0.39 lakh on July, 2011, as per the data submitted by NSE. These details give an impression that prior to the intervention of SEBI and other authorities the client code modification was being done in cases other than that for rectifying the genuine errors under the garb of existing practices.

17. I note that NSE has issued a circular ref: NSE/INVG/2011/670 dated August 26, 2011, in pursuance of SEBI’s circular ref: CIR/DNPD/6/2011 dated July 05, 2011. In the circular issued by NSE, it has stipulated specific criteria to identify the genuineness of the client code revisions. Thus, it is clear that NSE could have taken corrective measures such as those adopted by the instant circular, with the intervention of SEBI. Having been convinced that NSE has been negligent in discharge of its duties even though it might not have been a party to the mischief, I have to decide on the appropriate penalty to be awarded to NSE.

While dealing with matters concerned with discharge of regulatory functions, there would be few occasions where monetary penalty would be appropriate. Suspending/ interrupting the working of the stock exchanges is also not an appropriate penalty as it involves negative externalities and could be considered only in extreme cases. Therefore, given the nature of the lapses and the efforts made, I am of the opinion that penalty of warning would be appropriate in this case.

18. I, therefore, in exercise of the powers conferred upon me in terms of Sections 11(2)(j) and 19 of the Securities and Exchange Board of India Act, 1992 read with Section 12A of the Securities Contracts (Regulation) Act, 1956, hereby warn National Stock Exchange of India Limited to be more cautious and perceptive in discharge of its regulatory duties.

DATE : April 10, 2012 PRASHANT SARAN PLACE: MUMBAI WHOLE TIME MEMBER SECURITIES AND EXCHANGE BOARD OF INDIA