PRE-PUBLICATION VERSION

COMMODITY FUTURES TRADING COMMISSION

Concept Release: Performance Data and Disclosure for Commodity Trading Advisors and Commodity Pools

AGENCY: Commodity Futures Trading Commission

ACTION: Request for Comments

SUMMARY: The Commodity Futures Trading Commission ("CFTC" or "Commission") wishes to obtain public comment regarding possible changes to regulatory requirements which apply to the programs offered to the public by commodity trading advisors ("CTAs") and commodity pool operators ("CPOs"). The proposals discussed in this release originate from two sources. First, National Futures Association ("NFA") submitted a set of proposals (the "NFA Proposal") to the Commission for its approval, which concern computational and disclosure matters relating to participating in CTA programs on a partially-funded basis. Second, the Commission staff�s preliminary review of the NFA Proposal gave rise to a number of additional related proposals which the Commission also wishes to consider. The NFA Proposal is set forth separately in a section entitled "NFA Proposal," in the form in which it was submitted to the Commission for approval. NFA's and the Commission staff�s related proposals, collectively, fall within the following categories: 1) improving risk profile data for clients considering participation in CTA programs on a partially-funded basis, 2) providing CTA client account information to FCMs for risk management purposes, 3) improving risk profile data on commodity pools, 4) providing a theoretically sound basis of computation and presentation for rate of return ("ROR") and related risk profile data, 5) improving the presentation of historical performance and risk profile data, and 6) providing periodic statements of program activity and results to CTA clients.

All of the proposals, including the NFA Proposal and the additional proposals originated by the Commission staff, are discussed in detail in Part IV of this release, entitled "Request for Comment." At the end of each section, questions are posed to help focus public comment on the issues raised. Comment would also be welcome on any related issue and need not be limited to the questions posed in this release.

After considering the comments received, the Commission may approve or disapprove the NFA Proposal without further public notice, may request NFA to amend its proposal, or may propose for public comment changes to various Commission rules, advisories or interpretations pertaining to performance reporting and disclosure.

DATE: Comments must be received on or before __________, 1998. (Insert date to represent 60 days after date of publication in the FEDERAL REGISTER.)

ADDRESS: Interested parties should submit their comments to Jean A. Webb, Secretary of the Commission, Commodity Futures Trading Commission, Three Lafayette Centre, 1155 21st Street, N.W., Washington, D.C. 20581. Reference should be made to "Performance Data and Disclosure for Commodity Trading Advisors and Commodity Pools." In addition, comments may be sent by facsimile transmission to (202) 418-5221 or by electronic mail to [email protected].

FOR FURTHER INFORMATION CONTACT: Paul H. Bjarnason, Jr., Chief Accountant, (202) 418-5459, electronic mail: "[email protected];" Robert B. Wasserman, Special Counsel, (202) 418-5092, electronic mail: "[email protected];" Kevin P. Walek, Branch Chief, (202) 418-5463, electronic mail: "[email protected];" or Eileen R. Chotiner, Futures Trading Specialist, (202) 418-5467, electronic mail: "[email protected]," Division of Trading and Markets, Commodity Futures Trading Commission, 1155 21st Street, N.W., Washington, D.C. 20581.

SUPPLEMENTARY INFORMATION:

I. BACKGROUND

Past performance information presented to clients and prospective clients is a primary marketing tool for CTA programs and commodity pools. This type of information appears in disclosure documents, advertisements, promotional materials, and in compendia prepared by third-party services. Performance information is also reported either directly to clients to communicate the results of the CTA�s trading on behalf of their accounts or in periodic reports to investors in public and private commodity pools.

The Commission�s aim is that information provided to clients be accurate, complete, and understandable. The Commission believes that performance data can be useful to clients as a way of making risk and return comparisons among investment alternatives. Performance information can assist clients in distinguishing one CTA from another in terms of historical willingness to undertake risk, fee load, volatility and longer term results or facilitating comparisons with other investment opportunities. However, the Commission recognizes that requiring more data does not always result in better information for clients. It does not wish to overload clients with excessive amounts of data, nor does it wish to burden CTAs and CPOs with excessive requirements. As noted above, the Commission and NFA have identified ways to improve existing regulatory requirements that apply to CTAs and CPOs. This release discusses a variety of issues and requests public comment thereon.

II. DISCUSSION

  1. Rate-of-Return
  2. The Commission�s current requirements for the presentation of ROR data are based upon the "return on investment" ("ROI") concept used by economists, financial analysts and other professionals throughout the business world to measure the results of a variety of investment activities, from real estate development to internal capital budgeting to securities or commodities trading. ROI is used to compare various types of investments, as well as different investment managers. However, in all areas outside of commodities trading, the divisor used in the calculation of ROI represents an actual "investment" of tangible assets of the client � that is, the divisors used are amounts of actual cash funding that are owned or borrowed by the investor.

    ROR is calculated, in accordance with Commission regulations, by dividing the net performance by the beginning net asset value ("BNAV") as of the beginning of the period. Under current Commission advisories, the BNAV used to calculate the ROR must be based on a set of "fully-funded" accounts � accounts for which the "nominal account size" at the inception of the trading program is equal to the "actual funds" subject to the CTA�s access and control.

    "Actual funds" held pursuant to the CTA�s trading program are funds deposited with the client�s FCM either (1) in an account for which the CTA is granted discretionary trading authority or (2) in another account, subject to a binding agreement permitting the FCM to transfer funds to the first account at the direction of the CTA and committed to the CTA�s trading program, as demonstrated by factors specified in Advisory 87-2.

    Commission Rules 4.25 and 4.35 require that the performance of accounts directed by a CTA be disclosed for the past five years and the current year to date. In order to permit performance data to be disclosed without excessive detail and repetition, the rules permit the performance of all reasonably comparable accounts in each of a CTA's programs to be shown on a composite basis. When performance disclosure requirements were first adopted by the Commission over 20 years ago, the data required under the rules provided only a simple historical perspective on the profits earned or losses incurred by the participants in a CTA�s or CPO�s programs. However, in recent years the Commission has amplified the requirements to include data which provides a clearer focus on volatility, as opposed to simply displaying profits and losses. The performance capsules are now required to include, among other things, monthly rates of return for the most recent five calendar years and the current year-to-date, the worst monthly percentage drawdown during that time period, the worst peak-to-valley percentage drawdown for the time period, and the amount of funds under management.

  3. Nominal Account Size
  4. The "nominal account size" is an amount the CTA and the customer have agreed upon, usually in a written contract. It determines the level of trading for the client relative to other accounts in the CTA�s program, regardless of the level of actual funds. This means that customers of a given CTA who have the same nominal account size will have the same trades placed for their accounts. Generally, it also means that a customer who has agreed to a nominal account size of twice that of another customer of the same CTA will have twice the number of positions. The use of nominal account sizes simplifies management of the trading for a multiplicity of accounts, especially where the desired level of trading by the clients is not represented by the actual funding levels, as explained below.

    It is important to point out what nominal account size does not represent. It does not represent a particular number of positions, since there are times when a CTA may believe it prudent to stay out of the markets entirely or, alternatively, to be more aggressive than usual. It is not a function of margin requirements, nor is there any absolute or constant relationship to margin requirements arising from the CTA�s trading. While in a retail context, the nominal account size is sometimes described as an amount sufficient to make it unlikely that any further cash deposits will be necessary over the course of the client�s participation in the CTA�s program, the client may not look to the nominal account size as a maximum possible loss, since unexpected losses could exceed the nominal account size. Therefore, the nominal account size does not represent the limit of the customer�s liability, nor may any CTA represent that it is an indication of the maximum likely or possible loss that may be incurred.

    Nominal account sizes are not comparable from one CTA to the next. In discussions with representatives of the industry concerning this issue over the past ten years, it has become clear to the Commission staff that there is no method in common use in the industry relating the nominal account sizes to the number of positions traded. Indeed, NFA has reported that setting such levels "is inherently a subjective process" and "a matter of the CTA�s judgement."

    Nominal account size is sometimes referred to as a "legally binding" amount. While the amount specified does establish some legally binding obligations between the customer and the CTA, these only extend to (1) the basis of the management fees to be paid by the customer and (2) the trading level to be employed by the CTA for this account relative to other accounts managed under the same program. The nominal account size does not represent an obligation to furnish an amount of actual funds. The account arrangement between the CTA and its client may be terminated by the client at any time regardless of the amounts deposited in any account over which the CTA has or had trading authority. Of course, the client must settle any debits left in the account at the FCM as a result of trades ordered by the CTA before termination. As indicated above, these debits could exceed the nominal account size.

    The fact that nominal account size does not represent an actual investment � or even a commitment � of tangible funds and the lack of a commonly accepted method for determining the nominal account size have been major factors in the Commission�s reluctance to permit the use of the nominal account size in determining ROR, except as permitted by Advisory 93-13.

  5. Evolution of Present Commission Requirements

As mentioned above, the Commission�s requirements have evolved over time in response to identified problems and issues. One of the issues which has been at the forefront of consideration is the so-called "notional funds" issue. This issue pertains to the determination of the BNAV, which is the amount to be used as the divisor in the computation of ROR. The Commission first addressed this issue in 1987. Consistent with current Commission rules on the matter, Advisory 87-2 affirmed that only actual funds on deposit could be used in determining BNAV. Its purpose was to permit inclusion in BNAV of funds which are not carried at the FCM, but which can be reached by the FCM to satisfy a margin call. Advisory 87-2 provided that actual funds for the ROR calculation could include funds carried at the FCM or located at other depositories to which the FCM had access. This Advisory was needed because a literal application of the Commission�s rules resulted in the exclusion of some funding for accounts which logically should have been included. For the successful trader, the undue minimization of BNAV had the effect of resulting in unrealistically magnified RORs. The converse was true for losses. However, issuing Advisory 87-2 did not solve all of the reporting issues.

Some clients deposit to the account managed by a CTA actual funds which are only a fractional percentage of the nominal account size. This practice is referred to as "notional funding" or "partial funding." As indicated by NFA, the widespread use of partially-funded accounts raises the issue of how to report the performance of these accounts in a manner which is not misleading and without creating an undue number of performance tables. Prior to 1993, the Commission�s reporting scheme was entirely based on "actual funds."

Advisory 93-13�s main feature was the "fully-funded subset" method of ROR reporting. Under this method, the RORs presented in the performance table were not based upon all the accounts in a CTA�s program. The RORs were based only upon the fully-funded accounts � hence, the name "fully-funded subset" method. The Advisory provided for a matrix to permit clients to convert the fully-funded subset RORs to RORs for various partial funding levels. To qualify for the method, the fully-funded accounts must, in the aggregate, represent at least ten percent of the total nominal amount of funds traded by the CTA in the trading program. The Advisory also requires that the CTA make certain additional calculations to ensure that the subset is representative of the CTA�s program. As long as the two tests are met, this method produces approximately the same ROR as does a method (such as the NFA Proposal) that bases BNAV on the nominal account size.

The Commission has also sought to highlight the risk of CTA trading programs and commodity pools. In August 1995, the Commission enhanced requirements for the disclosure of the risk of volatility in all CTA and CPO programs by adding two new disclosure requirements � the largest percentage monthly drawdown and peak-to-valley drawdown for each program or pool offered by a CTA or CPO. The Commission felt that this new dimension to performance data provided a valuable heightened focus upon the risks of commodities trading, namely the possibility of large drawdowns of equity � either on a monthly or continuous basis.

Since August 1995, the Commission has received requests to address CTAs that have difficulty achieving the fully-funded subset necessary to qualify to use Advisory 93-13. The interest in this issue suggests that partially-funded account programs are becoming more prevalent. Because of the possibility that more clients are participating on a partially-funded basis, the Commission has become concerned that fully-funded basis data may be irrelevant or misleading for a growing segment of clients. Since partially-funded accounts are more highly leveraged than fully-funded accounts, they will incur magnified gains and losses compared to fully-funded accounts. For example, a customer who is funding its account at 25% of the nominal account size will realize gains � and losses � at four times the rate experienced by a fully-funded client. A loss of 30% on a fully-funded basis will result in a loss of 120% of the investment of a customer which funds its account at 25% of the nominal level, wiping out the initial investment and leaving a deficit to be repaid by the customer.

The Commission has also noted that commodity pools are accessing CTA programs on a partially-funded basis. Therefore, commodity pools raise similar concerns, because their disclosure documents contain information on the pool�s CTAs only on a fully-funded basis.

III. NFA PROPOSAL

On February 26, 1998, NFA submitted for Commission approval a change to its Compliance Rule 2-29(b)(5) that would require RORs for CTAs to be based on the nominal account size as described in proposed NFA Compliance Rule 2-34, rather than upon the actual funds which are associated with the CTA�s program, as presently required by Commission regulations. Proposed NFA Compliance Rule 2-34 and a related Interpretive Notice, both of which were previously submitted for Commission approval, specify certain requirements regarding account documentation and disclosure for partially-funded accounts, as well as certain disclosure requirements for CPOs. Together, the amendments to NFA Compliance Rule 2-29(b)(5), proposed NFA Compliance Rule 2-34, and the proposed Interpretive Notice constitute the NFA Proposal.

The NFA Proposal requires a CTA who directs a client�s account to enter into a written agreement with the client that includes:

1) The account size which the CTA will use as the basis for its trading decisions, i.e., the nominal account size;

2) The name or description of the trading program in which the client is participating;

3) Whether the client will deposit, maintain or make accessible to the FCM an amount equal to or less than the nominal account size; and

4) How additions, withdrawals, profit and losses will affect the nominal account size and the computation of fees.

The CTA would be required to provide a copy of this agreement to the FCM carrying the client�s account. The CTA would be required to disclose, in writing, the factors considered by the CTA in determining any minimum account size of the trading program in which the client is participating. In addition, unless a client is a qualified eligible client as defined in Commission Rule 4.7, the CTA would be required to disclose the following information in writing:

1) An estimated range of the amount of customer equity generally devoted to margin requirements or option premiums, expressed as a percentage of the nominal account size, and an explanation of the effect of partially funding an account at that percentage;

2) A description of how management fees will be computed, expressed as a percentage of the nominal account size, and an explanation of the effect of partially funding an account at that percentage;

3) An estimated range of the commissions generally charged to an account, expressed as a percentage of the nominal account size, and an explanation of the effect of partially funding an account at that percentage; and

4) A statement that the greater the disparity between the nominal account size and the amount deposited, maintained with or made available to the FCM, the greater the likelihood, and possible size, of margin calls.

The NFA Proposal prohibits the use of ROR figures in promotional material unless such figures are calculated in a manner consistent with that required under CFTC regulations and are based on the nominal account size as described in NFA Compliance Rule 2-34.

The NFA Proposal also imposes disclosure requirements on CPOs who allocate assets among the pool�s CTAs in such a way that the total allocations to its CTAs are greater than the total assets of the pool. In particular, the CPO must disclose the following information in writing to all participants except QEPs, as defined in Commission

Rule 4.7:

1) A statement of the total amount allocated to CTAs as a percentage of the pool�s net assets;

2) A description of how management fees charged by the CPO and the CTAs will be computed, including a statement of the total amount of management fees charged to the pool as a percentage of the pool�s net assets;

3) An estimated range of the amount of commissions and transaction fees that will be charged to the pool in the next twelve months and an estimate of these fees as a percentage of the pool�s net assets; and

4) A statement that allocating in excess of the pool�s net assets among CTAs has the effect of proportionately magnifying the profits and losses that may be incurred by the pool.

NFA presents several reasons for its Proposal. NFA states that basing BNAV solely on the amount deposited by the client with the FCM can distort the past performance results reported to clients. The accounts of two clients who have permitted the CTA to base its trade orders on the same account size during the same time period, using the same program, can show very different RORs based solely on their cash management strategies. According to NFA, this factor has nothing to do with the CTA�s trading decisions. NFA believes that a CTA�s performance history should reflect the results of the CTA�s trading decisions and should not be affected by the client�s cash management strategies. NFA further believes that computing ROR for partially-funded accounts based on actual funds on deposit overstates both positive and negative returns in those accounts. In addition, NFA believes that the fully-funded subset is so restrictive that more and more CTAs have been unable to use it.

NFA also recognizes that there are valid concerns regarding the documentation, disclosure, and sales practice problems that notional funding can create. According to NFA, however, these concerns are not computational issues to be addressed through BNAV but are separate issues that should be addressed independently of the ROR calculation. Therefore, NFA has proposed using the nominal account size for calculating BNAV and imposing the separate requirements, which are set forth above, to address these compliance concerns.

IV. REQUEST FOR COMMENT

The Commission shares NFA�s concern for accurate disclosure. In this connection, the proposals, collectively, are designed to ease the calculation of ROR for CTAs and enhance the amount and quality of data available to prospective clients of CTAs and investors in commodity pools. In considering the issues involved, the Commission wishes to obtain as much information as possible and to consider all relevant options. The sections below contain discussion and pose questions regarding several broad topic areas. The Commission does not wish to limit comment to the issues and questions set forth below, and comment is welcome on any aspect of CTA or commodity pool ROR reporting, accounting or disclosure.

  1. Disclosure of Risk Profile Data on CTA Programs For Clients Considering Participation on a Partially-Funded Basis

The Commission staff suggests consideration of expanded disclosure of historical percentage drawdown data, as explained below.

Discussion: Presently, drawdown data is required to be presented for CTA programs only on a fully-funded basis. The Commission staff has become concerned that historical drawdown data presented only on a fully-funded basis may mislead investors who are considering a partially-funded participation. It is important to convey to investors, as clearly as possible, that partially-funded participation in a CTA program will result in proportionately greater volatility � and proportionately greater drawdowns � compared to a fully-funded participation. Accordingly, the Commission wishes to explore the costs and benefits of requiring drawdown percentage data to be presented at two or three partial-funding levels that are representative of those offered by the CTA (e.g., at the 25%, 50%, and 75% levels) in addition to the fully-funded level. Presenting actual drawdown data on a partially-funded basis would illustrate the volatility of partial funding with a clarity that could not be achieved in a textual discussion. A CTA would not be required to present information for partial funding levels which are below the minimum offered by that CTA (e.g., a CTA which does not accept accounts which are funded at less than 50% partial funding would not be required to present information at the 25% level).

Questions:

1) What would be the costs and benefits of presenting drawdown figures geared to two or three partial funding levels?

2) What would be the most effective format for the presentation?

B. Presentation of Data Concerning Estimated Margin Ratios

NFA proposes to require CTAs to disclose, to any client which is not a QEC under Commission Rule 4.7 and which partially funds a participation in a CTA�s program:

an estimated range of the amount of customer equity generally devoted to margin requirements or option premiums, expressed as a percentage of the nominal account size of the accounts traded by the CTA, and an explanation of the effect of partially funding an account at that percentage.

Proposed Rule 2-34(b)(1) (emphasis added).

Discussion: This ratio, which is to be presented to partially-funded customers, is nonetheless a measure of the CTA�s program on a fully-funded basis, since it is based upon the nominal account size. It appears that use of the ratio is intended to provide a measure or indicator of the risk of the CTA�s program. The additional textual requirement is designed to help clients understand how partial funding increases such risk.

The Commission believes any new required disclosure should be assessed in light of its clarity, reliability in achieving its intended purpose, and its potential for being misunderstood by investors. If this proposed disclosure were required, it is possible that prospective clients will compare CTAs on the basis of this ratio. This possibility leads to the following issues for consideration:

Questions:

1) Will disclosure of information concerning the margin ratio, as discussed above, be useful to potential investors? Please give details of how potential investors will use this information.

2) What evidence, in the form of studies or otherwise, supports the proposition that margin requirements are a reliable indicator of the level of risk?

3) Does a requirement that CTAs disclose an "estimated" range of the amount of customer equity "generally" devoted to margin involve a standard so inherently discretionary that it creates a danger of presenting information that is misleading to potential investors?

4) Would a requirement that CTAs commit to an absolute maximum percentage of customer equity devoted to margin, beyond which no margin-increasing changes will be made, provide a more useful disclosure structure? What would be the advantages and disadvantages of such a structure? How should such a structure be implemented?

  1. Would any other alternative structures present more useful information? What would be the advantages and disadvantages of such structures?

C. Providing the CTA/Client Agreement to the FCM

The NFA Proposal calls for the CTA to provide a copy of the CTA/client agreement to the FCM carrying the customer�s account.

Discussion: NFA has indicated that it believes an FCM would find the nominal account size useful as a general indicator of the amount and size of trading intended to be undertaken in the account on behalf of the customer. The FCM could use this information in making a determination as to whether to accept this client and, if so, under what credit terms.

Questions:

  1. Do FCMs consider the client�s nominal account size useful information? Do they currently obtain such information? Would the imposition of a regulatory requirement aid them in doing so?
  2. Would a different method of providing the FCM with information concerning nominal account sizes be more efficient? What method (if any) of communication should be required? What should the timing and the form of this communication be?

D. Presentation of Risk Profile Data on Commodity Pools

The NFA Proposal imposes various disclosure requirements on CPOs that allocate assets among a pool�s CTAs in such a way that the total allocations to its CTAs are greater than the total assets of the pool. One of the requirements is for the CPO to provide a statement of the total amount of nominal account sizes allocated to a pool�s CTAs as a percentage of the pool�s net assets. The Commission desires to obtain comment on an alternative method of presenting a risk profile for a commodity pool which was developed by its staff.

Discussion: The most readily apparent use for NFA�s proposed ratio would be for prospective clients to compare one commodity pool to another. On initial consideration, it might seem that the greater the amount of the nominal account size compared to pool net assets, the greater the risk of a pool would be. But in this connection there are some issues that should be explored.

Although nominal account sizes may be useful in the context of an individual CTA, it does not follow that the ratio would be a consistent measure for even a single pool over time. As noted above, nominal account sizes are not comparable across CTAs. Therefore, a ratio based on the aggregate of nominal account sizes would not lend itself to making accurate and reliable comparisons between pools. Moreover, the ratio of one CTA�s nominal account size to the others may change over time. The Commission is interested in reviewing evidence which contradicts or supports this preliminary conclusion.

The Commission wishes to explore an alternative approach to enhancing the presentation of risk profile data for pools. This approach is founded on the precept that the volatility of a pool is a function of the volatilities of the investment vehicles (i.e., CTA programs or investee funds) in which it has invested. Therefore, the Commission wishes to consider requiring the presentation of data disclosing, on a pro forma basis, the effect of the worst historical drawdown for each of the vehicles the pool invested in over the course of the year. Such a presentation requirement might be implemented as follows:

1) For each investment vehicle selected, present the worst monthly and worst peak-to-valley drawdown percentages on a leveraged basis for:

a) the investment vehicle itself, at the pool�s leveraged basis (e.g., if the fully-funded worst drawdown for CTA "X" was 10 percent and the pool funds its participation in the program of CTA "X" on a 50 percent basis, the worst drawdown would be presented as 20 percent); and

b) the investment vehicle�s historical pro-forma impact on the pool, as though the highest percentage of pool assets over the past year were invested in the investment vehicle for the full historical period, at the leverage level of the pool (e.g., if CTA "X" had been allocated 25 percent of the pool�s net assets, the 20 percent worst monthly drawdown would be presented as a 5 percent impact (20% * 25 %) upon the pool�s net assets).

2) For major investee funds, data on the investee fund�s major investments would be required on a "look-through" basis, if they qualified as material under the selection criteria discussed below.

3) Finally, for each investment vehicle, identify the number of days during the year that the fund was invested in the vehicle and whether it is currently so invested.

An example of such a presentation follows:

Investment Investment (Leveraged) Highest Percentage of Fund Impact on Fund Number of Days Held
Worst Month Worst Peak-to-Valley Worst Month Worst Peak-to-Valley
CTA X (20%) (Y%) 25% (5%) (Y * 25%) 365

The purpose of the selection criteria is to select investment vehicles for which detailed risk profile data must be provided, i.e., those which expose the pool to the risk of material loss. It is also important to limit the number of vehicles for which information is presented, to avoid overwhelming the investor with an excessive volume of data. Finally, the criteria should consider the pool�s investments over the course of a year, rather than on a particular date, to avoid strategic behavior aimed at "cleaning up" the portfolio for a single measurement day. One example of a selection method would be the following:

Identify each investment vehicle in which, at any time during the course of the year, the actual funds invested by the pool equaled or exceeded five percent of the pool net assets. For each such investment vehicle, calculate an index which is the product of (A) the greatest amount invested (by notional value) times (B) the vehicle�s worst monthly drawdown percentage, times (C) the number of days during the year that the pool was invested in this vehicle. Present the data described above for the investment vehicles with the top N index values.

Questions:

1) What evidence supports or contradicts the proposition that the ratio between aggregate notional value and total pool net asset value is a useful measure of the risk level of a commodity pool?

2) Would presentation of leveraged worst drawdown data, as described above, for a selection of a commodity pool�s investment vehicles provide useful information to potential investors? What would be the disadvantages of providing such information? What is the most effective means of presenting such information? Should the results of the calculations described above be presented, or should different information be presented?

3) Are the selection criteria described above useful? Would a different selection method be more appropriate? For how many investment vehicles should the data be presented?

4) When should this table be presented: in disclosure documents? Sales literature? Pool annual reports?

E. Theoretical Soundness of the Basis of Computation and Presentation for ROR and Related Risk-Profile Data

The NFA Proposal does not require CTAs to maintain any fully-funded accounts to validate their nominal account sizes. By contrast, current practice, as described in Advisory 93-13, requires a fully-funded subset comprised of fully-funded accounts accounting for ten percent of the aggregate nominal account sizes, to validate the nominal account sizes. The Commission wishes to explore the implications of this change.

Discussion: The Commission has always sought to ensure that the methodologies it has required or permitted to be used in the various reporting schemes under its jurisdiction are based upon sound economic and accounting principles. In this connection, wherever possible, the Commission adheres to Generally Accepted Accounting Principles ("GAAP") in CTA, commodity pool, and FCM financial reporting. The fully-funded subset method permitted in Advisory 93-13 is consistent with the Commission�s historical approach to standards by requiring that the nominal account sizes set by the CTA be validated by the existence of a subset of accounts that are fully-funded with actual assets, pursuant to GAAP. This explicit linkage to actual funds, in effect, permitted the RORs to have some basis in traditional financial and accounting methods. By contrast, the NFA Proposal, which permits unrestricted use of the subjectively established nominal account size, lacks such an anchor or reference point.

Questions:

  1. Should the fully-funded subset requirement be retained to validate the nominal account sizes used by the CTA, or should it be dropped entirely?
  2. Does the fact that many CTAs may have difficulty in obtaining a fully-funded subset demonstrate a flaw in the regulatory methodology, or does it demonstrate an unrealistic setting of nominal account sizes? In other words, if the greatest actual funding level for any of a given CTA�s accounts was 50% (e.g., all $1 million nominal accounts are funded at $500,000 or less), is it not more accurate to express the nominal account sizes at 50% of their initial level?
  3. If the fully-funded subset should be dropped, what would be the theoretical basis for the method of computing ROR, in terms of economic and financial accounting theory?
  4. How do nominal account sizes used by CTAs generally fit into the broader world of financial services, so that a potential investor might fairly compare investments in commodity pools with other potential investments?

F. Changes in the Presentation of Historical Data

Current regulations require disclosure of approximately five years of historical ROR data, presented on a monthly basis, and presentation on a capsule basis of the single worst monthly drawdown and worst peak-to-valley drawdown during the same period. The Commission wishes to consider the costs and benefits of requiring a longer time-frame for disclosing performance data for CTAs and commodity pools while reducing the period for which disclosure of monthly data is necessary in the basic disclosure document. The focus of the disclosure document would be to provide key profile information. The Commission staff has also suggested that the Commission consider expanding the number of worst drawdown months presented, from one to three or possibly six. The overall effect of this change would be to reduce the number of data items presented in the disclosure document, while increasing the scope of the information made available to the investor.

Discussion: In many markets, extreme market events do not always occur within a five-year time-frame, which is the limit of the present requirement. Often the time interval between market events is ten years or more. Thus, limiting the historical presentation requirements to a five-year period, as the current regulations do, may permit some CTAs and commodity pools to omit their greatest drawdowns from their historical risk profiles. Requiring data for a longer period will present a fuller picture to prospective clients. Such disclosure is especially important where notional funding is used, given the magnification of drawdowns inherent in partial funding.

The Commission also seeks to strike a balance between the sometimes conflicting goals of requiring all data that would be useful and avoiding the presentation of a volume of data that is cumbersome to read and analyze or too complex or voluminous to be easily assimilated by the prospective client. Therefore, the Commission staff has suggested that the Commission consider reducing the number of years for which monthly data is required and presenting the balance of the information on an annual basis or on some other summary basis, as discussed below.

In connection with consideration of reducing the number of monthly data items, the Commission staff has suggested that the Commission consider requiring more detailed information concerning the volatility of the CTA�s program, either by requiring presentation of an expanded number of worst drawdown months, e.g., the three worst months or the six worst months, or by requiring presentation of the standard deviation of the monthly returns. Presently, only disclosure of the worst single monthly return is required. Given the unreliability of past performance data as a predictor of future performance and the relatively greater correlation between past and future volatility, presentation of data which is more indicative of volatility seems warranted.

Questions:

1) What are the costs and benefits of requiring performance data for a period greater than the past five years? What period should be required?

2) How many years of monthly data should be required? What would be the most effective method of presenting such data? What would be the most appropriate method of presenting data for earlier periods (e.g., annual performance, annual performance plus footnoted standard deviation of monthly performance, etc.)?

3) What data should be presented to enable investors to measure the volatility of returns from a CTA�s program or a commodity pool? How many months of worst drawdown data should be required (e.g., one, three, six)? What would be the most effective format for the presentation of this data?

G. Keeping Clients Regularly Informed Regarding CTA Program Status

The Commission seeks to ensure that clients receive timely and complete information on the status of their participation in CTA programs.

Discussion: Commission rules do not currently require that CTAs provide any periodic reports to their clients. Presently, the only information the Commission requires to be

reported to a client is that provided by the FCM (e.g., trade confirmations and monthly account statements provided to the CTA's clients and to the CTA). However, this information does not fully inform the customer as to the status of its participation in the CTA�s program. Among the items the customer may also need are the following: (a) account fees (e.g., the amount of fees earned/charged during the period, payments received from client on amounts owed during the period both through charges to the client account at the FCM and from other sources outside the FCM account, and any balance unpaid by or credit due to the client at the end of the period); (b) information on the basis of incentive fee calculations (including the amount of unrecovered prior losses carried forward); and (c) the current nominal account size (i.e., amount originally agreed to, changes during the period and balance at end of period). It also may be useful to require the monthly statement to contain the management and incentive fee percentages, even though they are contained in the CTA/client agreement. This would permit the clients more easily to verify the amounts charged.

Questions:

1) Which of the data items discussed above would be valuable for clients to receive on a regular basis from CTAs? Are there any other data items which should be required? How often should this information be reported to clients? Is there a particular format which should be required?

2) What would be the costs for CTAs to report this information to clients on a regular basis?

3) On balance, what reporting requirements, if any, should be established?

V. CONCLUSION

The Commission believes that it is appropriate to examine concerns regarding ROR computation and other performance issues which are raised in connection with the proposals made by the Commission staff and NFA. The Commission hopes to develop a balanced approach to address these issues that will enable performance data provided to customers to be as useful and meaningful as possible, while not being excessively burdensome to CTAs and CPOs. To this end, the Commission requests public comment on the proposals and the related issues set forth above.

Issued in Washington, D.C. on June 11, 1998 by the Commission.

____________________________________

Jean A. Webb

Secretary of the Commission

APPENDIX I

COMPLIANCE RULES

* * *

RULE 2-29. COMMUNICATIONS WITH THE PUBLIC AND PROMOTIONAL MATERIAL

* * *

(b) Content of Promotional Material.

No Member or Associate shall use any promotional material which:

* * *

(5) includes any specific numerical or statistical information about the past performance of any actual accounts (including rate of return) unless such information is and can be demonstrated to NFA to be representative of the actual performance for the same time period of all reasonably comparable accounts and, in the case of rate of return figures, unless such figures are calculated in a manner consistent with that required under CFTC Rule 4.25(a)(7)(i)(F) and are based on the nominal account size (as described in Compliance Rule 2-34).

* * *

Rule 2-34. DIRECTED ACCOUNTS AND COMMODITY POOLS

(a) At the time a Member CTA enters into an agreement to direct a client's account, the Member CTA must obtain a written agreement signed by the client (or someone legally authorized to act on the client's behalf) which states:

(1) the account size which the CTA will use as the basis for its trading decisions, i.e., "the nominal account size";

(2) the name or description of the trading program in which the client is participating;

(3) whether the client will deposit, maintain or make accessible to the FCM an amount equal to or less than the nominal account size, i.e., to fully or partially fund the account; and

(4) how additions, withdrawals, profits and losses will affect the nominal account size and the computation of fees.

The Member CTA must provide a copy of the agreement to the FCM carrying the account. The Member CTA must also disclose in writing the factors considered by the CTA in determining any minimum account size of the trading program in which the client is participating.

(b) Unless the client is a qualified eligible client under CFTC Rule 4.7, any Member CTA which directs a partially funded account must provide the following information in writing to the client:

(1) an estimated range of the amount of customer equity generally devoted to margin requirements or options premiums expressed as a percentage of the nominal account size and an explanation of the effect of partially funding an account on that percentage;

(2) a description of how the management fees will be computed, expressed as a percentage of the nominal account size and an explanation of the effect of partially funding an account on that percentage;

(3) an estimated range of the commissions generally charged to an account expressed as a percentage of the nominal account size and an explanation of the effect of partially funding an account on that percentage;

(4) a statement that the greater the disparity between the nominal account size and the amount deposited, maintained or made accessible to the FCM, the greater the likelihood, and possible size of, margin calls.

(c) Unless the pool participants are qualified eligible participants under CFTC Rule 4.7, any Member CPO which allocates assets among the pool's CTAs in such a way that the total allocations to its CTAs is greater than the total assets of the pool must provide the following information in writing to the pool participants:

(1) a statement of the total amount allocated to CTAs as a percentage of the pool's net assets;

(2) a description of how management fees charged by the CPO and the CTAs will be computed, including a statement of the total amount of management fees charged to the pool as a percentage of the pool's net assets;

(3) an estimated range of the amount of commissions and transaction fees which will be charged to the pool in the next twelve months and an estimate of such fees as a percentage of the pool's net assets; and

(4) a statement that allocating in excess of the pool's net assets among CTAs has the effect of proportionately magnifying the profits and losses which may be incurred by the pool.

(d) Each CTA Member which directs accounts and each CPO Member which allocates assets among CTAs in such a way that the total committed is greater than the total assets of the pool shall maintain the records required by this Rule in the form and for the period of time required by CFTC Rule 1.31.

(e) Each CTA Member which directs accounts and each CPO Member to which this rule applies allocates assets among CTAs in such a way that the total allocated is greater than the total assets of the pool shall establish and enforce adequate procedures to review all records made pursuant to this Rule and to supervise the activities of its Associates in complying with this Rule.

INTERPRETIVE NOTICE

NFA COMPLIANCE RULE 2-34

The Board of Directors recently passed NFA Compliance Rule 2-34, Documentation and Disclosure for Partially Funded Accounts. The Board recognized that certain customers may, for their own legitimate business purposes, deposit with the FCMs carrying their accounts less than the amount which they have directed the CTA trading their account to use as the basis for trading decisions. The Board sought to ensure that in such situations performance records accurately reflect trading results, that there is an adequate audit trail to verify past performance records and that customers receive adequate disclosures on the implications of partially funded accounts.

In the Board's view, the solicitation of partially funded accounts, particularly with less sophisticated customers, raises a number of compliance issues. Therefore, the Board wishes to make clear that NFA Compliance Rule 2-34 does not in any way diminish a Member's responsibilities under other NFA rules, most notably NFA's sales practice rules, when dealing with a customer who is considering a partially funded account.

Specifically, the Member must ensure that any solicitation present a balanced view of the risks and benefits of such an arrangement and disclose all material information. Furthermore, under NFA Compliance Rule 2-30, the Member must obtain the specified information regarding its customer's experience and financial condition and, in light of that information, must provide the customer with an adequate description of the risks of his investment. As the Board stated in its Interpretive Notice of that rule, for some customers the only adequate disclosure is that futures trading is simply too risky for that customer. That is particularly true when retail customers are induced to increase their leverage further by partially funding a trading account.

Any Member soliciting unsophisticated customers to trade with a partially funded account will bear the burden of demonstrating that its solicitation was in compliance with all NFA requirements.