New York Mercantile Exchange February 29, 2000

RULE ENFORCEMENT REVIEW

OF THE

NEW YORK MERCANTILE EXCHANGE

I. INTRODUCTION � PURPOSE AND SCOPE

The Division of Trading and Markets ("Division") has completed a rule enforcement review of the trade practice surveillance and disciplinary programs of the NYMEX Division of the New York Mercantile Exchange (�NYMEX� or �Exchange�) for compliance with Sections 5a(a)(8) and 5a(b) of the Commodity Exchange Act ("Act") and Commission Regulation 1.51.1 In addition, the Division assessed the Exchange�s audit trail and recordkeeping systems for compliance with Commission Regulation 1.35. This review covers the period of March 1, 1999 to February 29, 2000 ("target period").²

In conducting its review, Division staff reviewed Exchange documents that included, among others, computer reports and other documentation used routinely in conducting trade practice surveillance; selected trade practice investigation and disciplinary action files;³ investigation, inquiry, recordkeeping, audit trail, disciplinary, and floor surveillance logs; minutes of the disciplinary committee meetings; trading card and order ticket reviews; and compliance manuals and guidelines. In addition, Division staff interviewed senior Compliance Department (�Compliance�) officials, including the Vice President of Compliance, the Compliance Counsel, and the Director and Associate Director of Trade Practice Surveillance.⁴

The Division provided the Exchange with the opportunity to review and comment on a draft of this report on September 8, 2000. On September 14, 2000, Division staff conducted an exit conference with NYMEX officials to discuss the report�s findings and recommendations.

II. AUDIT TRAIL AND RECORDKEEPING - SECTION 5a(b) AND COMMISSION REGULATION 1.35

A. Essential Trade Data � Section 5a(b)(1)

Section 5a(b)(1)(B) of the Act requires that an exchange�s audit trail and recordkeeping system be able to capture essential data on the terms, participants, and sequence of transactions, including relevant data on unmatched trades and outtrades. To meet this requirement, NYMEX requires that each member sequentially record both customer and proprietary trades at time of execution on an integrated trading pad.⁵ The seller then must submit the �pit card� within one minute of the transaction by throwing the card into a netted area in the center of the pit where an Exchange employee retrieves and timestamps the card.⁶ This timestamp represents the one-minute execution time required pursuant to Commission Regulation 1.35(g).⁷

After the card is timestamped, the trade data are entered into TMS by an Exchange keypuncher. Within approximately 4 to 10 minutes, the pit card data appear on both the buyer�s and seller�s computer screens for validation and entry of the remaining required data for clearing, including customer identifier, customer type identification (�CTI�) code,⁸ and clearing member identifier⁹. The completed data are used for clearing purposes and become the basis for the Exchange�s automated trade monitoring system (�TXN�)¹⁰ and trade register, referred to as the �streetbook.�

Even if the members disagree with respect to the terms of the trade, under the single-side trade data entry system used on NYMEX, once a pit card is submitted, the trade is �locked in� and guaranteed at the time the Exchange staff inputs trade data from the card. The trade will clear, based upon the seller�s terms, even if one or both members fail to enter the remaining required clearing data. As a consequence, outtrades and unmatched trades are virtually nonexistent at NYMEX.

Based upon the foregoing, the Division believes that the NYMEX audit trail captures essential data on trades, including data on unmatched trades and outtrades, as required by Section 5a(b)(1)(B) of the Act.

B. One Minute Timing Compliance � Section 5a(b)(2)¹¹

The Exchange assesses the timeliness of pit card timestamps through the Pit Card Validation Report, which compares pit card times to trade times appearing on the Price Change Register (�PCR�) for all trades except spreads.¹² The Pit Card Validation Report reflects all outright sell transactions for each member, the number and percentage of such transactions which are valid and invalid and, for those that are invalid, the number of minutes that the pit card is late. The report deems a trade to be valid when the PCR shows a print of the trade price at a point during the minute before, the minute of, or the minute after the actual timestamp on the transaction�s pit card.

To assess the accuracy of the Exchange�s pit card times, the Division reviewed the Pit Card Validation Report for two randomly selected days during each of the 12 months within the target period. The daily percentage of compliance with the Exchange�s one-minute timing standard ranged from 89 percent to 92 percent, with an overall average accuracy rate of 91 percent.

To enforce the requirement that pit cards be timely submitted, the Exchange assesses the pit card timing accuracy of all trading members monthly with the Pit Card Validation Report. Any member whose monthly submission average for valid trades falls below 80 percent for the month is subject to graduated sanctions.¹³ During the target period, the Exchange levied 24 fines of $5,000, nine fines of $2,500, 12 fines of $1,000, eight fines of $750, 17 fines of $500, and 26 fines of $250, for a total of $175,000, and issued 189 warning letters. Compliance also forwarded four cases to the BCC for further action.¹⁴ In addition, in order to determine if late submissions are due to any type of trade practice violation, Compliance periodically initiates trade practice inquiries or investigations into the trading activities of members who consistently fail to meet the 80 percent threshold. Eight such investigations were opened during the target period.¹⁵

The Division believes that NYMEX has an adequate program for reviewing and enforcing compliance with its one-minute timing requirements. This is supported by the significant level of sanctions imposed during the target period. The Division also believes, however, that the Exchange�s program would be further enhanced by the inclusion of spread trade times in the Pit Card Validation Report such that the report could include spreads in its trade timing accuracy calculations. Therefore, the Exchange should expedite the planned program changes designed to include spreads in its trade timing accuracy calculations.

C. Recordkeeping: Floor Order Tickets and Trading Cards - Commission Regulation 1.35(a-1)(2)-(4) and (d)

1. Order Tickets¹⁶

To evaluate member compliance with Exchange Rule 6.18, which sets forth order ticket recordkeeping requirements, Compliance reviews order tickets during the course of trade practice investigations and order ticket reviews. A review consists of an examination of a minimum of 30 hand-written floor order tickets from one broker group per month, alternating monthly between the Comex and NYMEX Divisions.¹⁷ A broker group is most often selected for review based upon the appearance of the order tickets examined during the course of an investigation.¹⁸ Each floor order ticket is examined for the required account identification and timestamps. All filled or partially filled orders are required to have entry and exit timestamps; unfilled or canceled orders are required to have at least one timestamp; and orders for which the terms have changed are required to have a timestamp which corresponds to the change.¹⁹

Upon completion of the floor order ticket review, a branch order ticket review is initiated, if corresponding branch order tickets were created.²⁰ Compliance selects order tickets associated with one futures commission merchant (�FCM�) for which the broker group fills orders. Compliance then conducts a review similar to that performed for floor order tickets for between 10 and 15 corresponding branch order tickets. In addition, Compliance reviews the branch order tickets to determine if the terms of the orders and customer account numbers/designations correspond to those on the floor order tickets and compares the timestamps on both sets of order tickets to the associated pit card times and time and sales prints in order to assure that they all correspond. Broker groups and FCMs found not to be in compliance with recordkeeping requirements are subject to disciplinary action.²¹

The Division reviewed the six Exchange order ticket reviews completed during the target period and examined copies of order tickets within the files. The Division�s review revealed that the Exchange conducted thorough reviews. The Exchange examined a total of 492 floor order tickets, of which 428 (87 percent) reflected account identification and 452 (92 percent) contained appropriate timestamps. The Exchange also examined ten branch order tickets, all of which reflected account identification and appropriate timestamps, and order terms and other data that corresponded to the associated floor order tickets. As a result of the six reviews, NYMEX issued three staff warning letters to three broker groups.²² The Division believes that NYMEX has adequate procedures for reviewing and enforcing compliance with its order ticket recordkeeping requirements.

2. Trading Cards²³

To evaluate member compliance with Exchange Rule 6.90, which sets forth trading card recordkeeping requirements, Compliance conducts a routine review of three days worth of each member�s original trading cards at least once annually.²⁴ Based upon scores derived from review of the trading cards for compliance with eight categories included in a NYMEX checklist,²⁵ members are determined to be in �Full Compliance,� in �Effective Compliance,� or �Not in Compliance.�

A member with no discrepancies in any of the eight categories would be in �Full Compliance;� a member with a compliance rate of 90 percent or better (increased from 85 percent on July 19, 1999) in categories one through three, no more than two violations in category four, and no discrepancies in categories five through eight, would be in �Effective Compliance;� and a member who failed to meet these standards would be �Not in Compliance.� Members found to be in the latter category are subject to disciplinary action. A warning letter is issued for the first finding of �Not in Compliance.� A second infraction within a 12-month period results in a $100 fine; and a third infraction within 18 months results in a $500 fine. A fourth infraction within 24 months results in a referral to the BCC for formal disciplinary action.

NYMEX completed 560 of the 714 trading card reviews it initiated during the target period. As a result of the 560 completed reviews, NYMEX issued 227 Full Compliance letters, 195 Effective Compliance letters, and 71 staff warning letters. The Exchange also assessed 26 fines of $100 and 25 fines of $500, for a total of $15,100. In addition, Compliance forwarded 16 cases to the BCC for further disciplinary action.²⁶

The Division reviewed 24 trading card reviews that resulted in warning letters or fines during the target period and examined copies of a sample of trading cards within those files. The Division also examined the 16 trading card reviews forwarded to the BCC. The Division�s review revealed that the Exchange conducted thorough reviews according to schedule. During the course of the 40 reviews that resulted in warning letters or disciplinary action, Compliance examined 1,731 trading cards. Of these, 841 cards (49 percent) failed to include a manual time of execution next to the first transaction on the card; 158 (nine percent) contained unused lines that were not marked through; and 16 (one percent) did not identify the open or close. One trading card reflected a skipped line. The Division believes that NYMEX has adequate procedures for reviewing and enforcing compliance with its trading card recordkeeping requirements.

D. Conclusions and Recommendations

The Division found that NYMEX maintains an adequate program for reviewing and enforcing compliance with its one-minute timing requirement for outright trades. The daily percentage of compliance with the Exchange�s one-minute timing standard as measured on the Pit Card Validation Report ranged from 89 percent to 92 percent, with an overall average accuracy rate of 91 percent. The Division also found that the progressive nature and amounts of fines imposed by the Exchange are adequate to encourage member compliance with the one-minute timing requirement. The Division believes, however, that NYMEX�s trade timing accuracy calculations would be enhanced by the inclusion of spread trade times, which the Exchange is in the process of implementing.

The Division also found that the Exchange has adequate procedures to assess and enforce member compliance with recordkeeping requirements. During the course of six order ticket investigations closed during the target period, the Exchange examined a total of 492 floor order tickets and ten branch order tickets. With respect to the floor order tickets, 87 percent reflected account identification and 92 percent contained appropriate timestamps. The Exchange found that all ten of the branch order tickets reflected account identification and appropriate timestamps and that the order terms and other data on the tickets corresponded with that on the associated floor order tickets. With respect to trading cards, NYMEX conducts a routine review of three days worth of each member�s original trading cards at least once annually. During the target period, NYMEX initiated 714 trading card reviews and completed 560.

At this time, the Division has no recommendations in this area.

III. TRADE PRACTICE SURVEILLANCE � SECTIONS 5a(a)(8) AND 5a(b) AND COMMISSION REGULATIONS 1.51(a)(2), (4), (5) AND (6)²⁷

A. Compliance Staff

Compliance, which is responsible for detecting, investigating, and prosecuting potential trading violations on both the NYMEX and Comex Divisions, is directed by the Vice President, Compliance. The trade practice staff is managed by the Director, Trade Practice Surveillance, who has overall responsibility for the investigation and prosecution of rule violations. The Trade Practice Surveillance staff includes an Associate Director, who assists the Director in the day-to-day management of the trade practice surveillance program, three managers, one supervisor, nine analysts, two clerks, and one floor observer.

The managers and the supervisor oversee the investigations and provide guidance and direction to the analysts and clerks. The nine analysts conduct, among other things, reviews of trading activity and trading documents, analyses of computerized exception reports, floor surveillance, and investigations of potential rule violations.²⁸ Each analyst is expected to open one inquiry per month.²⁹ The clerks perform administrative functions and conduct certain audit trail review procedures. The floor observer conducts floor surveillance and serves as a contact point between Compliance and the Exchange membership. Finally, the Trade Practice Surveillance unit is supported by one Compliance Counsel and two Associate Counsels. Based upon the foregoing, the Division believes that the Exchange has adequate staff to develop investigations and prosecute disciplinary matters.

B. Floor Surveillance

In order to detect trading violations, Compliance staff routinely observes trading in each contract market on the open and close, at a random time during the trading day, and when special market conditions warrant. The one full-time floor observer surveys activity on the floor throughout the trading day. In addition, two analysts are assigned to observe the opening, one analyst is assigned to observe the markets for at least 20 minutes during the middle of the day, and all analysts who are available observe the daily close of the markets.³⁰ According to the Exchange�s investigation logs, two inquiries and three investigations into possible trade practice violations were opened during the target period based on floor surveillance observations.

C. Electronic Surveillance

The Exchange�s primary surveillance program used to monitor trading and to detect and assist in the investigation of trade practice abuses is its computerized surveillance system, referred to as TXN.³¹ TXN is a highly flexible system that permits analysts to manipulate default parameters to create customized reports that focus on particular types of trading violations, members, or suspected patterns of violations. TXN was designed to be integrated with other surveillance tools and routinely uses Exchange data such as the PCR, streetbook, trade adjustments, and broker association database, to provide for up-to-date analysis of Exchange trading activity.

TXN provides two standardized surveillance screens that are reviewed daily by the analysts: the trading ahead and prearranged trading reviews. The trading ahead program displays instances where a broker appears to have traded ahead of either his or her own customer order or the customer order of an associated broker. The prearranged trading report displays instances where a broker may directly or indirectly have taken the opposite side of either his or her customer order or the customer of an associated broker. Both reports display any potentially profitable offsetting trades for each displayed exception. Both reports have preset default timing parameters such that trades with valid pit card times are evaluated with respect to one another if they fall within a two-minute window and trades with invalid pit card times are compared if they fall within a five-minute window.

In addition to the standardized daily screens, Compliance makes extensive use of TXN�s �Analysis� feature, the pattern recognition ability of TXN. Analysis detects patterns of trading activity among one or more brokers that suggest possible abusive practices with respect to trading ahead, prearranged trading, and �fairness� of fill prices.³² Analysis examines all cleared data on a quarterly basis and calculates the number of times that sequences of trades involving the same brokers present similar sets of fact patterns. Collections of such fact patterns that are statistically significant from a trade practice surveillance perspective are grouped by TXN into �leads.� Thus, leads suggest trading relationships between brokers statistically based on the patterns of trading in which they appear to be engaged.

In addition to the quarterly generation of leads, Analysis examines each day�s trading activity to determine if any new trades fit the pattern identified in the lead. If they do, a tally is presented in a summary screen to show the analyst that the identified pattern appears to be continuing. Analysts refer to leads during the course of daily reviews, as well as in connection with inquiries and investigations, to determine whether expansion of the inquiries or investigations to include other relevant parties and patterns is warranted. Analysts also open inquiries and investigations based upon leads. The Analysis menu is also being developed to pick up fact patterns reflecting profits and losses among members.

Other TXN programs include �Groups� and �Profiles.� Groups looks at trading activity among two or more brokers in an attempt to identify unregistered broker associations. Profiles is a statistical compilation that provides, for a three-month period, the distribution of any broker�s trading activity among CTI codes. It will also identify the top five opposite brokers with whom the broker trades and reflect the distribution of those trades by CTI codes. Finally, with respect to cross trades, the Exchange has a program that looks at apparent cross trades that were not identified as such and Compliance periodically reviews cross trades to assure such trades are executed competitively. As with other TXN programs, the Exchange�s broker association database is integrated into the review program such that TXN identifies instances where a broker crosses his or her own account with the account of either his or her own customer or the customer of an associated broker.

D. Adequacy and Timeliness of Investigations

1. Adequacy

During the target period, Compliance opened 43 inquiries and 123 investigations.³³ With respect to inquiries, according to the Exchange�s logs, 18 of the 43 inquiries (42 percent) were internally generated by Compliance based upon review of TXN screens or the streetbook and seven (16 percent) were opened based upon trading card reviews. The remaining 18 inquiries were opened for a variety of reasons that included, among others, floor observations (two), NFA reports (two), and anonymous, broker, or customer complaints (seven). Of the 41 inquiries closed during the target period, nine resulted in the opening of investigations and one was referred to market surveillance staff.

With respect to investigations, 46 of the 123 investigations (37 percent) were internally generated by Compliance based upon review of TXN trading ahead, prearranged trading, or cross trade screens or the streetbook.³⁴ Additionally, 42 investigations (34 percent) were audit trail or recordkeeping-related;³⁵ 15 (12 percent) were opened based upon customer complaints; and the remaining 20 were opened for a variety of reasons that included, among others, NFA reports or registration issues (five), floor observations (three), Commission referrals (two), and trading on ACCESS (two). Seven of the 143 investigations had previously been opened as inquiries.

Forty-one inquiries and 108 investigations were closed during the target period. Of those investigations, 25 resulted in the referral of members to the BCC for disciplinary action and 31 resulted in the issuance of staff warning letters. In addition, two were trading card reviews that resulted in summary fines of $100 and $500, one was an anonymous complaint that was closed and subsequently opened as an inquiry, and one was closed with the administration of a verbal advisory. Forty-eight investigations were closed with no further action recommended.

The Division reviewed for adequacy and completeness 16 of the 41 inquiries and 81 of the 108 investigations closed during the target period, including all 16 of the closed investigations that involved customer complaints.³⁶ The Division also reviewed six investigation files that were closed prior to the target period because these files corresponded to disciplinary cases that became final actions during the target period. The files reviewed included, among others, investigations of possible trading ahead of or against customer orders, prearranged trading, and other noncompetitive trading that were opened based on information obtained from a variety of sources. They also included trading card, order ticket, cross trade, and late pit card submission reviews.

The Division found that the Exchange conducted thorough, well-documented investigations and made appropriate analyses. Investigations generally were expanded in scope, as appropriate, to review additional instances of possible violative activity and to examine the records of additional parties. Investigation files contained underlying trading documents, summaries of witness interviews, correspondence, computer reports, and investigative and activity logs. Files for cases forwarded to the BCC included detailed investigation reports that discuss the facts of the case and give staff�s conclusions and recommendations. Files for investigations closed with no further action to be taken were, for the most part, equally complete. Staff decisions to close these reviews without disciplinary action or referral were supported by adequate analyses.

The Division found one investigation, however, that it believes should have been expanded to include a more detailed review of additional instances of possible prearranged trading. More importantly, the Division believes that this investigation, even without being expanded, when taken together with a series of related investigations and inquiries, appears to form the basis for a referral to the BCC for consideration of charges.

On December 19, 1997, Compliance opened an inquiry to examine possible instances of prearranged trading among four brokers identified during review of the Exchange�s computerized surveillance program.³⁷ The inquiry led to investigation 050-98, opened August 31, 1998, which examined three of the four brokers based upon a possible pattern of disclosure and noncompetitive and prearranged trading. In the investigation, Compliance reviewed several instances of possible prearranged trading, but elected to reconstruct only one instance from one trade date in detail. In the reconstruction of trading records from the one trade date, staff found an apparent pattern that involved a high concentration of trades between the broker�s customer orders and the two other brokers� personal accounts and a high concentration of customer sales being executed at the low and buys being executed at the high on the open. Staff also found inconsistencies in the sequencing of trades and alterations made to prices and quantities on all three members� trading cards. Nonetheless, Compliance concluded that there was no reasonable basis to conclude that any Exchange rules were violated. Despite similar findings, these conclusions were also reached in a customer complaint that was investigated in conjunction with investigation 050-98.³⁸

During the course of investigation 050-98, on January 25, 1999, staff opened an additional inquiry based upon review of the TXN prearranged trading program to examine four instances of questionable personal trading activity between two of the three brokers.³⁹ This inquiry led to another investigation, opened on January 27, 1999, in which staff concluded that there appeared to be prearranged and noncompetitive trading.⁴⁰ Staff also concluded, however, that despite the appearance of noncompetitive trading, there was insufficient evidence to support the allegation at the BCC level. Staff issued warning letters to both brokers.

Another inquiry involving the same three brokers was opened on May 11, 1999, shortly before the close of investigation 050-98.⁴¹ This inquiry was opened to examine another possible pattern that involved a high concentration of trades between one of the broker�s customer orders and the two other brokers� personal accounts, and a high concentration of customer sales at the low and buys at the high on the open. Staff determined that although there again was a high concentration of trades executed on the high and low on the open, the trades were executed with various opposite brokers. Thus, staff concluded that there was insufficient evidence to conclude that Exchange rules were violated.

Although the Exchange was persistent in reviewing these members� trading activity, the Division believes that the Exchange should have taken additional steps with respect to this sequence of events. First, Compliance should have expanded the scope of investigation 050-98 in order to reconstruct in detail more than one apparent instance of prearranged trading identified by the Exchange�s computerized surveillance system. In light of the possibly violative trading patterns found by Compliance, as well as the alterations in sequencing and in prices and quantities on trading cards, staff should have looked at all trading records for each of the three members for at least several, if not a majority, of the identified instances.

Second, based on the continued findings of a high concentration of trades between one broker�s customer orders and the other two brokers� personal accounts and a high concentration of customer sales at the low and buys at the high on the open, the Division believes that the investigations and inquiries, when viewed together, contain adequate evidence of possible noncompetitive trading such that Compliance should have forwarded the results to the BCC for a determination as to whether charges were appropriate. Further, one of the three brokers was the subject of a disciplinary action during the target period for, among other things, noncompetitive trading, prearranged trading, trading ahead of his customers, and taking the opposite side of his customer orders.⁴²

Third, the Division found no indication that the members reviewed were either fined for recordkeeping violations or selected for out-of-sequence trading card reviews. Recording trades out of sequence is a violation of NYMEX Rule 6.90(B) and Commission Regulation 1.35(d)(2). Corrections to data recorded on trading cards must be made in accordance with NYMEX Rule 6.90(E) and Commission Regulation 1.35(d)(7).

2. Timeliness

The Division found that the majority of the Exchange�s investigations were completed in a timely manner in accordance with Commission Regulation 8.06.⁴³ Based on NYMEX�s investigation and customer inquiry logs, the Exchange closed 108 investigations during the target period, including 30 opened prior to the target period. Of the 108 closed investigations, 64 (59 percent) were closed within four months. This represents a significant improvement over the finding in the 1996 Review that 37 percent of the investigations were completed in four months.⁴⁴ An additional 14 investigations (13 percent) were closed between four and six months, and another 21 investigations (19 percent) were closed between six months and one year.⁴⁵ Nine investigations were closed within 14 months.

According to the Exchange, there are three factors generally responsible for the length of time taken in the nine longest investigations. First, some of the investigations were very fact intensive and required numerous reconstructions and/or requests for and review of additional trading documents. Second, Compliance experienced staff turnover within the trade practice group, which resulted in the reshuffling and reassignment of investigations. Third, significant delays resulted when factors outside of normal operations required the efforts of the trade practice staff, such as staff�s response to the problems caused by the extremely active and volatile trading in the COMEX gold options market on September 28, 1999.

The Exchange opened a total of 123 investigations during the target period, of which 78 (63 percent) were closed during the target period. Of the 45 investigations opened during the target period that remained open at the end of the period, 25 (56 percent) had been open for four months or less, nine (20 percent) had been open for between four and six months, and 11 (24 percent) had been open for between six months and one year.

E. Conclusions and Recommendations

The Division found that the Exchange maintains an adequate trade practice surveillance program. The Exchange�s investigations, including those initiated in response to customer complaints, were thorough, generally expanded in scope to include additional trading sequences and members where appropriate, well documented, and completed in a timely manner. Interviews with members were conducted where appropriate, investigation files contained relevant documentation, and investigation reports complied with the content requirements of Commission Regulation 8.07. The Division also found that NYMEX has improved its performance with respect to the closure of investigations within the four-month timeframe set forth in Commission Regulation 8.06.

Therefore, based on the foregoing, the Division recommends that NYMEX:

� Present to the Business Conduct Committee for its consideration the results of related investigations and inquiries which, when taken together, appear to contain evidence that a reasonable basis exists to find that Exchange rules may have been violated.

IV. DISCIPLINARY PROGRAM � SECTION 5a(b) AND COMMISSION REGULATION 1.51(a)(7)46

A. Sanctions Imposed

The Division reviewed the Exchange's Disciplinary Register, various disciplinary committee minutes, Regulation 9.11 disciplinary action notices reflecting disciplinary actions taken during the 12-month target period, investigation files for all investigations forwarded to the BCC during the target period, and disciplinary files for all cases resulting in disciplinary action that became final during the target period. During that period, the Exchange took final disciplinary action against 18 members and two clerks in 17 cases.⁴⁷

Cases finalized during the target period, most of which were resolved through settlement agreements, resulted in fines totaling $364,500. Of this fine total, $351,000 (96 percent) was assessed in cases involving trade practice and/or floor trading offenses. NYMEX also issued a permanent ban from the floor, two permanent bans and one three-year ban from the handling of customer orders,⁴⁸ suspensions of 120 days (one), two weeks (two), and one week (two), and eight orders to cease and desist in the cases involving these offenses. The remaining $13,500 in fines assessed through disciplinary action were issued in nine settlement agreements with members who were found to have committed a fourth infraction of the Exchange�s trading card rules within a 24-month period.⁴⁹ In addition to the fines assessed in disciplinary actions, the Exchange issued $190,600 in fines under summary disciplinary procedures for audit trail and recordkeeping violations.

The largest monetary and other sanctions were assessed in a single case in which two members and a floor manager/phone clerk were disciplined for numerous rule violations related to their involvement in a scheme to defraud customers.⁵⁰ One member was fined $50,000, suspended for 120 days, and permanently banned from handling customer orders in any capacity for, among other things, having an interest in and operating a bucket shop, trading for himself opposite a customer transaction, failing to maintain records, and refusing to cooperate with a Compliance investigation. The second member was fined $5,000 and banned for three years from handling customer orders in any capacity for, among other things, noncompetitive trading, trading ahead of a customer order, disclosure of a customer order, prearranged trading, accommodation trading, and violation of a cease and desist order. The floor manager/phone clerk was fined $200,000 and permanently banned from the trading floor and from holding electronic trading privileges for, among other things, operating and making transactions for a bucket shop.⁵¹

Another trade practice case that resulted in significant sanctions was initiated with an investigation generated by TXN surveillance programs. A member charged with, among other things, trading ahead of customer orders and violation of the Exchange�s trading card recordkeeping rule entered into a settlement agreement that included a fine of $35,000, a suspension of two weeks, and an order to cease and desist.⁵² Compliance also found that the member reallocated trades to benefit his personal account to the detriment of his customer.

The Division found that the sanctions imposed during the target period were appropriate in light of the rule violations committed. In addition to assessing substantial fines, the Exchange imposed several bans and suspensions. However, the Division believes that in all settlements and disciplinary decisions where the Exchange can determine the amount of customer losses attributable to members� abusive trading activity, as it did in four cases finalized during the target period, the Exchange should order members to pay restitution to injured customers.⁵³

Two cases in which the amounts of customer losses were determined involved the same member. In the bucketing case, the Appeal Panel affirmed the findings and conclusions of the Hearing Panel that one member violated Exchange rules prohibiting, among other things, noncompetitive and prearranged trading with respect to two trading sequences. The member was fined $5,000 and banned for three years from handling customer orders. Compliance had calculated that the member disadvantaged his customer by $3,130 in those two trading sequences.⁵⁴ In another case involving the same member, the Appeal Panel upheld the Hearing Panel�s findings that the member violated Exchange rules prohibiting trading ahead and violated a cease and desist order in several sequences charged. The member was fined $7,500 and assessed a lifetime ban from handling customer orders. Compliance had calculated that the member was responsible for customer losses of at least $9,660 in those trading sequences.⁵⁵

Compliance also determined the amounts of customer losses in two other cases. In the case, discussed above, in which the member agreed to a fine of $35,000, a suspension of two weeks, and an order to cease and desist for trading ahead of executable customer orders and violation of the Exchange�s trading card recordkeeping rule, Compliance found that the member�s reallocation of contracts from a customer account to his personal account resulted in losses to the customer of $14,300.⁵⁶ Finally, in a case in which the member agreed to a fine of $7500, a two-week suspension, and an order to cease and desist for, among other things, trading ahead of his customers, Compliance found that the member�s trading ahead resulted in losses to customers of $368.⁵⁷

In addition to ordering members to pay restitution to injured customers, the Division believes that the Exchange�s disciplinary program could be improved by requiring that a member�s prior relevant disciplinary history be disclosed when the case is presented to any Exchange disciplinary committee for consideration of sanctions. Such information is appropriate for review when determining the severity of any sanction. Currently, NYMEX provides the member�s prior disciplinary history only as it relates to a Compliance request that the BCC consider issuing a charge that the member has violated a prior cease and desist order. Similarly, unless the member has been charged with violation of a cease and desist order, or unless the panel requests the information, the Hearing Panel adjudicating the member�s case may not be aware of his or her prior disciplinary history when determining what sanctions may be appropriate.

B. Timeliness of Disciplinary Procedures⁵⁸

Compliance routinely provided copies of investigation reports to BCC Panel members approximately one week prior to BCC meetings. A review of the Exchange's Disciplinary Register and BCC minutes indicates that BCC Panels met six times during the target period to consider investigations of NYMEX Division members and reviewed seven trade practice-related investigations involving eight members and three clerks⁵⁹ and 20 audit trail or recordkeeping-related investigations involving 17 members.⁶⁰ In each instance, the BCC made a determination as to whether a rule violation may have occurred and whether to charge a party at that first meeting.⁶¹

The Division also reviewed the amount of time that elapsed between the date of the BCC decision to issue a notice of charges and the date the complaint was served, as well as the amount of time which elapsed between the date of the issuance of charges and the date of a settlement or hearing. The Division found that Compliance generally issued the complaint within one to two weeks after the BCC meeting at which the complaint was authorized in audit trail and recordkeeping-related cases and within one to two months in more complex cases.

During the target period, the BCC Panel agreed to accept offers of settlement from four members.⁶² In each instance, the complaint was issued on the same day or subsequent to the date the settlement agreement was accepted by the BCC. With regard to the 16 settlement offers accepted by a Hearing Panel, the elapsed time between the issuance of the charge and acceptance of the settlement offer ranged from less than two months to nine months for the four members involved in trade practice-related cases and from less than one month to three months for members involved in audit trail or recordkeeping-related cases.⁶³

The Division also reviewed the timeliness of decisions rendered by adjudication panels in light of the requirements of Commission Regulation 8.18 and NYMEX Rule 8.11.⁶⁴ Two hearings were conducted during the target period. One decision was released in approximately two and a half months and one was released in just less than six months. The latter case involved a 15-count complaint and allegations relating to 14 separate trading sequences. The Division believes that the length of time taken to issue the decisions was not excessive under the circumstances of the cases. Thus, NYMEX generally appears to resolve its disciplinary cases in an expeditious manner.

C. Conclusions and Recommendations

Based upon its review, the Division found that the Exchange maintains an adequate disciplinary program. Disciplinary matters are promptly referred to disciplinary committees, findings in cases that proceed for further action appear to be supported by the evidence, penalties appear reasonable relative to the conduct being sanctioned, and disciplinary action is taken in a reasonably timely manner. Penalties imposed by NYMEX disciplinary committees included fines totaling $364,500 and a permanent ban from the floor, two permanent bans and one three-year ban from the handling of customer orders, suspensions of 120 days (one), two weeks (two), and one week (two), and 17 orders to cease and desist.

The Division also found, however, that although Compliance generally calculated customer losses attributable to trading abuses, the Exchange did not order members to pay restitution to injured customers. The Division believes that restitution should be included in all settlements and disciplinary decisions where customer harm can be determined to ensure that injured customers are appropriately compensated for losses due to a member�s abusive trading activity.

In addition, the Division believes that the Exchange should require that a member�s prior relevant disciplinary history be disclosed when the case is presented to a disciplinary committee for consideration of sanctions. Such information is appropriate for review with respect to determining the severity of any sanction.

Therefore, based on the foregoing, the Division recommends that NYMEX:

� Order restitution in all settlements and Exchange disciplinary decisions where the amount of customer harm can be determined.

� Disclose the member�s prior relevant disciplinary history when the case is presented to any Exchange disciplinary committee for consideration of sanctions.

1 This review does not include the Comex Division of the NYMEX. Rule enforcement reviews prepared by the Division are intended to present an analysis of an exchange's overall compliance capabilities for the period under review. These reviews deal only with programs directly addressed in the review and do not assess all programs. The Division's analyses, conclusions and recommendations are based, in large part, upon the Division's evaluation of a sample of investigatory cases and other exchange documents. This evaluation process, in some instances, identifies specific deficiencies in particular exchange investigations or methods but is not designed to uncover all instances in which an exchange does not address effectively all exchange rule violations or other deficiencies. Neither is such a review intended to go beyond the quality of the exchange's self-regulatory systems to include direct surveillance of the market, although some direct testing is performed as a measure of quality control.

2 The Division's previous rule enforcement review of the Exchange's compliance program was presented to the Commission on July 25, 1996 ("1996 Review"). In the 1996 Review, the Division found that NYMEX generally had adequate market surveillance, audit trail, trade practice surveillance, and disciplinary programs. The Division had deferred initiating a rule enforcement review of NYMEX since 1996 because of numerous changes undertaken at the Exchange, including the Exchange�s move into a new facility in mid-1997 and implementation of the Trade Management System (�TMS�) in early 1999. In addition, the Division conducted audit trail tests of the Exchange�s most active markets in 1998 and had been actively examining the Exchange�s trade monitoring system as a result of NYMEX�s September 30, 1998 pending petition for exemption from the prohibition on dual trading. The Division subsequently determined that it was appropriate to initiate the current review of the Exchange�s core compliance programs. A copy of the 1996 Review can be found in Appendix 1.

3 Division staff reviewed all disciplinary files for cases that became final during the target period. With respect to other investigations reviewed, Division staff selected a sample that included both investigations closed with no further action and investigations forwarded to disciplinary committees. The sample included investigations of various types of possible substantive trading violations as well as recordkeeping reviews.

4 The transcripts of April 11 and 13, 2000, can be found in Appendix 2.

5 The integrated pad consists of two sets of documents, the trading card and the pit card. The top document on the pad is the trading card, which consists of three soft-ply (one original and two copies), color coded sheets, with spaces for up to nine trades. The bottom document consists of four individually numbered hard-copy cards (pit cards) which are virtual duplicates of the trading card. All of the documents in the group contain the same unique preprinted sequence number. When trade information is recorded on the trading card, that information is concurrently recorded on the uppermost pit card. When a pit card is thrown into the pit, trade information subsequently recorded on the trading card is duplicated onto the then-uppermost pit card. As a result, all of the trade data recorded on the trading card is also recorded on one or more associated pit cards.

6 The pit card (and trading card) reflects essential trade matching data, including the commodity, date, identity of the executing and opposite broker, contract month, and price (or, in the case of spreads, the differential). For option trades, the pit card also reflects the premium and a put/call indicator. In addition, with the exception of trades executed during the opening or closing ranges, Exchange members are required to manually record the execution time of the first trade made on each trading card.

7 Members are also required to throw a pit card into the center of the pit when the last transaction recorded on the integrated pad was a buy transaction. This pit card is normally timestamped, but the timestamp is not entered into TMS. However, the pit card can be used to assist in the reconstruction of trading.

8 The CTI is a numerical code, required by Regulation 1.35(e), which is used to identify the source of a trade: CTI 1 designates a trade by a member for his or her personal account or an account over which the member has discretion; CTI 2 is a trade for his or her clearing member�s house account; CTI 3 is a trade for another member present on the floor, or for an account controlled by such member; and CTI 4 is a trade for any other type of customer.

9 NYMEX Rule 9.04(M) requires that the remaining data required for clearing must be submitted to the Exchange by the executing member within one hour after the trade information on the pit card appears on the member�s computer screen. Failure to enter timely all required data, referred to as an �incomplete transfer,� could result in summary disciplinary action administered by Compliance. If a member fails to submit transfer data on 10 or more trades, he or she will receive a warning letter. Within a six-month period following the warning letter, failure to submit such data on 10-20 trades will result in a $250 fine; a similar failure with respect to 20 or more trades will result in a $500 fine. A member who fails to submit transfer data on three or more occasions during a six-month period could be referred to the Business Conduct Committee (�BCC�) for possible issuance of a complaint. During the target period, the Exchange issued ten warning letters for failure to submit required transfer data.

10 TXN (now Data Ventures) was the name of the New Mexico company with which NYMEX contracted to develop the trade surveillance system.

11 Section 5a(b)(2) of the Act requires that the audit trail system of each exchange accurately record the times of trades in increments of no more than one minute in length and the sequence of trades for each floor broker and trader. In addition, Commission Regulation 1.35(g) requires that the trade execution time be obtained and stated in increments of no more than one minute in length on an exchange�s trade register. As previously noted, to meet these audit trail requirements, NYMEX requires that each seller complete a pit card and throw it into the center of the pit within one-minute of trade execution.

12 Spreads comprise approximately 25-35 percent of total NYMEX volume. NYMEX has a time and sales for two-legged spreads, which constitute approximately 80 to 90 percent of the total spread volume. The Exchange is currently in the process of compiling specifications for program changes to include two-legged futures spreads in its automated trade timing accuracy calculations. See NYMEX letter to the Division dated July 31, 2000, which can be found in Appendix 3.

13 Warning letters are issued for the first three monthly failures to meet the threshold. For subsequent offenses, progressively increasing fines are imposed, except that a floor broker who does not dual trade who attains a valid submission rate of 79 percent will be issued an �advisory letter.� A member is cited each month his or her audit trail percentage falls below 80 percent, except that members who experience a six month violation-free period will be granted a clean slate for purposes of a subsequent offense. The fine schedule is as follows: $250 for the first failure to meet the threshold following the third warning letter; $500 for the next violation; $750 for the next violation; $1000 each for the next three violations; $2500 each for the next three violations; and $5000 each for the next three failures to meet the 80 percent threshold. Each subsequent violation will result in another $5000 fine and a referral to the BCC with a recommendation that a complaint be issued for violation of Exchange Rule 8.55(B)(1), Conduct Detrimental to the Exchange, as a minor offense.

The Exchange increased the maximum sanctions and added referral to the BCC for consideration of charges as an additional disciplinary action in response to a Division recommendation in the 1996 Review.

14 Two members were each referred twice to the BCC. The BCC voted to issue a complaint in each instance.

15 Four of the eight investigations were closed during the target period. Three were closed with no action taken and one resulted in the issuance of a staff warning letter for failure to comply with the Exchange�s trading card recordkeeping requirements.

16 Pursuant to Commission Regulation 1.35(a-1)(2)(i), each exchange member or its designated person receiving a customer�s order on the floor must prepare a written record of the order in nonerasable ink, including the account identification and order number. The member then must timestamp the order with the �entry� time to reflect the date and time, to the nearest minute, that it is transmitted to or received on the exchange floor. When the execution price is reported from the floor, the order must be timestamped again, pursuant to Regulation 1.35(a-1)(4), to reflect the �exit� time.

17 A broker group on NYMEX is a business entity composed of one or more brokers who execute customer orders and are paid their transaction fees collectively through the Automatic Transfer of Money system.

18 In conducting an investigation, the analyst obtains the member�s trading records for the entire day. If there are deficiencies in any of the order tickets that are critical to determining whether a rule violation occurred, the member�s broker group is automatically issued a warning letter and is scheduled for a full order ticket review. During the target period, the Exchange issued four such warning letters. If the analyst observes deficiencies in the order tickets generally, the group may be scheduled for a review, depending upon, among other things, the date the group was most recently reviewed.

19 The analyst uses a detailed checklist that requires that he or she test for, among other things, carbon copies with account numbers written in ink, which is considered evidence that the account number was obtained after the order was filled. A copy of a checklist used in these reviews can be found in Appendix 4.

20 Comparatively few floor order tickets have corresponding branch order tickets. Excluding machine-generated order tickets routed via order routing systems, approximately 80 to 90 percent of the orders received on the floor at NYMEX are called in directly by customers to floor brokers.

21 A finding of 90 percent or better compliance results in no action. Compliance of 80 to 89 percent, in either the first or second review, results in the issuance of a staff warning letter. Compliance of 79 percent or below results in the following: a BCC warning letter and a follow-up review within six to eight months following the first review and referral to the BCC for the issuance of a complaint following the second review.

22 One broker group was issued a warning letter for failing to meet the threshold for both account identifiers and time stamps.

23 Commission Regulation 1.35(d) requires that trading members record information for all transactions on trading cards or other records, including the member�s name, opposite member�s identification, clearing member�s name, date, execution time, commodity, quantity, price and delivery month. In addition, members must record purchases and sales in nonerasable ink, in exact chronological order of execution, on sequential lines of the trading card without skipping lines between trades, and must cross out any remaining lines on the trading card. Errors on cards may be corrected, provided that the originally recorded information may not be obliterated or otherwise made illegible. Trades made during an exchange�s opening and closing periods must be separately identified. Regulation 1.35(d) also requires that trading cards contain pre-printed identification information and sequence numbers to distinguish one member�s cards from another�s, to permit the sequencing of cards, and to differentiate each card prepared by a member from such other cards for no less than a one-week period.

24 Compliance typically selects a target week and examines the trading cards of approximately 60 members, 10-20 of whom are selected based upon receipt of a warning letter or fine as a result of a previous review. The remaining 40-50 members are selected from a monthly computer report as part of the effort to assure that each member is reviewed at least once annually.

25 The eight categories include: 1) time recorded next to the first trade on each card; 2) identification of the open or close and marking through unused lines; 3) buys and sells recorded sequentially in chronological order without skipping or sharing lines; 4) drawing a single line through erroneous information; 5) using nonerasable ink; 6) cards used are maintained by the floor member; 7) cards are used in numerical sequence day-to-day; and 8) cards that are not used, or are rewritten by the floor member, are maintained. A copy of the checklist can be found in Appendix 5.

26 Nine of the 16 cases were finalized during the target period. Each of the nine members agreed to a fine of $1,500 and a cease and desist order.

27 Section 5a(a)(8) of the Act requires each exchange to enforce all bylaws, rules, regulations, and resolutions made or issued by it or by the governing board or any committee. Section 5a(b) of the Act requires each contract market to maintain and use a system to monitor trading to detect and deter violations of the contract market�s rules committed in the making of trades. Under Section 5a(b)(1), such a system must include the commitment of resources necessary for a trade monitoring system to be effective in detecting and deterring trade practice violations, including adequate staff to develop and prosecute disciplinary actions; trade practice surveillance systems capable of reviewing, and used to review, trade data to detect violations committed in making trades; and floor surveillance.

In addition, Commission Regulation 1.51 requires that each exchange use due diligence in maintaining a continuing program for the surveillance of trading practices on the floor of the exchange; for the investigation of customer complaints and other alleged or apparent violations of the exchange bylaws, rules, regulations, and resolutions; and for such other surveillance, record examination, and investigation as is necessary to enforce exchange bylaws, rules, regulations, and resolutions.

28 Based upon level of experience, from least to most, the nine analysts include one junior analyst, five analysts, one experienced analyst, and two senior analysts.

29 The distinction between an inquiry and an investigation is that the latter involves a request for documents that are not already under the control of the Exchange. In an inquiry, only documents available through Exchange sources including, among others, the streetbook, time and sales, and pit cards would be reviewed.

30 The full-time floor observer records his observations hourly on the Daily Floor Observation Report. Analysts assigned to conduct floor surveillance annotate daily observations on a Weekly Floor Observation Sheet. Sample copies of the floor observation reports can be found in Appendix 6.

31 The TXN system performs surveillance on trades executed on the floor and on ACCESS, the Exchange�s after-hours screen-based electronic trading system. The same data fields that are available through TXN for trading during regular trading hours are available for trading on ACCESS.

32 �Fairness� helps to identify situations where one broker may be getting an advantage on trades over other brokers, indicating that the broker may be involved in some sort of relationship with one or more members that is resulting in profits.

33 NYMEX�s trade practice analysts routinely open inquiries and investigations into instances of possible trading abuses. Inquiries consist of reviewing TXN and other trading data and documentation already in the possession of the Exchange. Inquiries may be initiated based on, among other things, questionable trading activity identified by TXN, floor observations, member referrals, or customer complaints. If an inquiry is opened and it is determined that no further action is required, the inquiry would be closed and a brief, closeout memorandum would be prepared. If the data reviewed during the course of an inquiry indicates that trading abuses may have occurred, the inquiry is closed and an investigation is opened. During investigations, relevant original trading records are requested and examined and relevant persons are interviewed. Investigations are also opened directly when information obtained indicates that a possible violation of Exchange rules may have occurred. Where daily surveillance reviews reveal particularly strong evidence of possible violative trading, an investigation may be opened immediately to obtain and examine relevant documents. In addition, a Commission or NFA referral generally results in the opening of an investigation. Copies of the Exchange�s investigation and inquiry logs can be found in Appendix 7.

34 Of the 46 investigations generated from TXN screens or the streetbook, 21 involved review of cross trades that may have been executed in violation of Exchange rules, 16 focused on potential trading ahead, seven involved possible prearranged trading, and two were directed at possible violation of the Exchange�s post-close trading rules.

35 Of these 42 investigations, 20 were opened because it was determined, during the course of a trading card review, that the member had been found not in compliance with the Exchange�s trading card rules for the fourth time within 24 months; 13 were opened because the member had received at least three warning letters for failing to submit pit cards in a timely manner; and eight represented order ticket reviews.

36 As previously noted, Division staff selected a sample that included both investigations closed with no further action and investigations forwarded to disciplinary committees. The sample included investigations of various types of possible substantive trading violations as well as recordkeeping reviews.

37 Inquiry 073-97.

38 03-98CI was opened on March 25, 1998, based upon two sources: an anonymous complaint that alleged possible prearranged trading among the same three members and a customer who complained that his sales on the open were the only sales executed at the low. Compliance found that the customer�s orders comprised the majority, but not all, of the volume at the low. Compliance also found, however, that the broker executing the orders had made price changes to all three of the executions allocated to the customer, in addition to quantity and price changes to other executions on the same trading card. Staff also identified an apparent pattern that involved a high concentration of trades between the broker�s customer orders and the two other brokers� personal accounts and a high concentration of customer sales at the low and buys at the high on the open, often accompanied by price changes on trading cards and trades recorded out of sequence. Nonetheless, as with investigation 050-98, staff concluded that there was no reasonable basis to conclude that any Exchange rules were violated.

39 Inquiry 008-99.