December 14, 1998

Margaret H. McFarland

Deputy Secretary

Securities and Exchange Commission

450 Fifth Street NW

Washington, D.C. 20549

Re: NASD Proposed Guidance on Mark-up and Mark-down Practices for Debt Securities Other than Municipal Securities; File No. SR-NASD-97-61.

Dear Ms. McFarland:

The Federal Regulation Committee and Self-Regulation Committee ('Committees") of the Securities Industry Association ("SIA") appreciate this opportunity to comment on proposed National Association of Securities Dealers ("NASD") Rule IM-2440-2 ("Proposed Interpretation"). The Committees believe that the Interpretation is in many respects a very helpful explication of how the existing NASD mark-up policy applies to transactions in government and other debt securities except municipal securities ("covered debt securities"). Attempting to clarify some aspects of mark-up and mark-down restrictions ("mark-up policy") with regard to covered debt securities is a worthwhile effort. A more coherent mark-up policy is important to strengthening public trust in the securities markets. With some important modifications noted below, the Committees believe that interpretive guidance can be extremely helpful. By giving proper recognition to the role of dealers in risking their capital to provide customers with liquidity, such guidance could improve investor protection and public confidence in these markets.

INTRODUCTION

From its inception, mark-up doctrine has been an anomaly in securities regulation - the imposition of government price regulation on securities transactions. The justifications for regulating mark-ups have been called increasingly into question over the past two decades, as the world of fixed rate commissions has given way to a much more competitive system. As a result, government "rate-making" via mark-up regulation looks increasingly out of place and out of date. Because of these fundamental changes in the nature of both the equity and fixed income markets, the Committees believe that the entirety of regulatory mark-up policy is ripe for reconsideration. Public confidence in many securities markets depends greatly on the willingness of dealers to provide liquidity to otherwise illiquid markets. Dealers are unlikely to extend capital to provide liquidity unless they can obtain a return commensurate with the risks to their capital. It is therefore critically important for mark-up policy not to deter dealers from risking their capital to provide liquidity in markets where it is most needed.

The benefit to investors of mark-up policy as applied to debt markets is particularly questionable. As we explain below (and as the Proposed Interpretation recognizes to some extent), unlike most equity securities, debt securities of different issuers that have similar credit and other performance characteristics are fungible with each other ("similar securities" in the parlance of the Proposed Interpretation).

Investors in these markets, who are generally institutions, can therefore readily assess the fairness of pricing based on yield to maturity, which is apparent on transaction confirmations. Because investors care about yield, and can use yield comparisons among similar securities to check on pricing fairness, controlling a dealer's mark-up or mark-down on a particular security adds little if anything to investor protection in these markets.

While holding our view that dealer-compensation regulation is in many respects anachronistic, and in some markets could be counterproductive to investor protection by impeding liquidity, the Committees believe that the effort to clarify the application of mark-up policy toward covered debt securities could be very helpful. The Proposed Interpretation correctly states that a mark-up is the difference between the price that the dealer, acting as principal, charges (or pays to) the customer and the prevailing market price for the security. This is consistent with our understanding that NASD and SEC mark-up policy are intended to limit the difference between the price set for customers and the prevailing market price, rather than to regulate a dealer's spread between bid and offer quotations, or to limit the dealer's profit based on original purchase price or ultimate sale price of the security. However, the Committees think that the Proposed Interpretation relies too heavily on equity-based characteristics of mark-up doctrine that do not adequately reflect fundamental differences in the roles of dealers in equity and fixed income markets. In particular, the Proposed Interpretation reflects too much of the fairly mechanical approach to assessing mark-ups developed in the context of equity securities. It is unrealistic to apply an equity-based approach to mark-ups in debt securities for three reasons. First, unlike most equity securities, different debt securities typically trade at net prices that reflect a prevailing market price among similar debt securities. Second, unlike many equity securities, most debt securities trade in markets that are overwhelmingly institutional in nature. Third, for many types of covered debt securities (such as high-yield and emerging market securities), the markets are substantially less liquid than the markets for most equity securities, resulting in greater risk to dealers willing to use their capital to facilitate customer trades. We offer several suggestions below for modifications to the proposal that should better tailor mark-up guidance to the debt markets.

1. In At Least One Respect, the Proposed Interpretation Expressly Goes Beyond Its Stated Intention Not to Create New Mark-Up Policy.

The Proposed Interpretation states that the appropriate mark-up or mark-down for most types of covered debt securities "is usually substantially less than 5 percent." Although the Proposed Interpretation repeatedly eschews any intention to expand on existing mark-up principles, on this point the Proposed Interpretation appears to go well beyond existing mark-up doctrine. NASD Conduct Rule IM-2440 currently provides that mark-ups in excess of 5% above prevailing market price may be unreasonable. We are not aware of any regulatory pronouncement, and of few litigated cases, where mark-ups below 5 per cent were deemed unreasonable. Such a sweeping pronouncement, applied across most of the highly variegated fixed income markets, is inappropriate and inconsistent with the NASD's statement that the Proposed Interpretation "is not intended to represent a departure from Rule IM-2440 .. . , but is being proposed to more accurately apply existing principles to government securities and other debt securities."

2. Mark-Up Doctrines Developed Primarily in the Context of Equity Securities Should Be Applied With Caution Toward the Debt Markets.

The Committees agree with the NASD that "debt and equity markets often differ in the extent and availability of inter-dealer transaction prices for a particular security. Notwithstanding these differences, the proposed guidance is founded on the premise that "the fundamental principles that are applied to mark-ups in equity securities apply also to the debt markets." The Committees believe that great caution should be exercised in applying equity-based mark-ups doctrine to fixed income markets.

There are tremendous degrees of variation between different segments of the debt markets. Generalizations about the fixed income markets are therefore difficult to make. However, one conspicuous and basic distinction between equity and many debt markets is that, while equity securities are typically valued by markets as determined by supply and demand based largely or entirely on the economic performance of the particular issuer, fixed income securities can be valued in relation to similar debt securities of other issuers.

This similarity among classes of fixed income instruments results in most covered debt securities trading in distinctly different ways from equity securities. Equities generally are either traded in centralized auction markets, or in electronic dealer markets that are functionally similar to auction markets. In contrast, most fixed income securities trade in decentralized OTC markets. Moreover, there are a far greater number of individual debt securities, and dealers who make markets in debt securities often do so for a wide range of debt securities with similar credit and yield characteristics, rather than focusing on providing continuous quotations in any one security.

Because of the tremendous variation in fixed income securities, there are enormous differences in the characteristics of markets for various types of debt securities. While much government debt and some types of corporate debt securities have highly liquid markets with fairly ready access to prices and quotations, an enormous number of debt securities, both government and corporate, trade in very illiquid markets lacking frequent price quotations, with many securities not trading on a daily, or even weekly, basis. Dealers play an essential role in providing customers such liquidity as exists in many fixed income securities, and assume substantial risks by contributing capital to provide that liquidity to otherwise illiquid markets. However, the dealers' role in these markets is unlikely to fit four-square within the definition of "market-making" as developed in the equity markets and articulated in the Proposed Interpretation. Without the ability to obtain adequate compensation for the capital that dealers in such securities put at risk, dealers are likely to withdraw from such markets, further impairing such liquidity as exists and limiting the opportunities for capital formation, particularly for young companies and for special-purpose entities raising capital through asset-backed offerings.

3. The Definition of "Market-Making" Should Be Expanded.

Under the Proposed Guidance, "integrated dealers . . . are permitted to calculate their mark-ups from contemporaneous sales prices to other dealers." The Proposed Guidance indicates that "integrated dealers" are dealers who sell to retail customers and also act as wholesale market makers. A "market maker" is defined as "a dealer who, with respect to a particular security, furnishes bona fide competitive bid and offer quotations on request and is ready, willing, and able to effect transactions in reasonable quantities at his or her quoted prices with other broker-dealers."

The Committees believe that this description of the circumstances under which a dealer should be permitted to calculate mark-ups or mark-downs on the basis of contemporaneous inter-dealer transactions fails to adequately recognize the ways in which dealers risk their capital to provide liquidity in many markets. Specifically, the definition is flawed in four respects. First, it could be read, probably unintentionally, as only applicable to dealers who sell to retail customers. Second, it does not sufficiently address dealers who act as market makers in a group of similar securities.

Third, it does not give adequate recognition to dealers who give quotations and stand ready to effect transactions with institutional customers on a regular basis. This is especially significant since inter-dealer transactions are rare in many debt securities,

but dealer-institutional customer transactions are much more common. Fourth, the proposal in its current form may exclude some dealers who make markets in illiquid securities. In light of the obviously greater risk of making market in illiquid securities that have infrequent or non-existent inter-dealer activity, points three and four suggest that the Proposed Interpretation in its current form could impair liquidity in many sectors of the debt markets.Retail Requirement. The Proposed Interpretation could be read to suggest that, in order to obtain the benefit of calculating mark-ups or mark-downs on the basis of contemporaneous inter-dealer quotations, a dealer must be an "integrated dealer," meaning that it must both sell to retail customers and act as a wholesale market maker. The Proposed Interpretation does not offer any explanation as to why interaction with retail customers should be a relevant consideration in determining whether a dealer is entitled to calculate mark-ups on a basis other than contemporaneous cost. Many dealers act as market makers for securities that are traded exclusively by institutional investors. We do not know of a reason why market makers in institutional markets should be disfavored over market makers in markets that have retail participation. We suspect that the Proposed Interpretation did not intend to suggest that a dealer must integrate retail activity with wholesale market-making in order to be able to rely on contemporaneous sales prices to establish prevailing market price. However, a clarification of this point would be helpful.

Market Makers for Groups of Similar Securities. The definition of "market maker" is unrealistically constricted to analyzing a dealer's activities "with respect to a particular security." The narrative discussion in the release accompanying the Proposed Guidance notes that "this language recognizes that dealers in debt markets may act effectively as market makers in a group of securities without publishing continuous two-sided quotations for each security within the group." Consistent with the NASD's own commentary, the definition of "market maker" in the Proposed Guidance should encompass activities regarding covered debt securities with similar credit and yield features.

Quotations and Transactions With Institutional Customers. Most fixed income securities trade in overwhelmingly institutional markets. It is not unusual for virtually all dealer transactions in a given debt security to be with institutional customers, rather than other dealers. This is because the enormous number of debt securities issuances often makes it economically infeasible for more than one dealer to do business in a particular issue. However, the similarity of groups of debt security issuances provides important market discipline - a dealer cannot survive as a market maker if its prices with institutional customers depart from those of similar securities offered elsewhere. Restricting the definition of market maker so that it looks only to inter-dealer quotations and transactions - neither of which may exist -- therefore fails to capture the nature of market making in many or most debt markets.

Given the sophistication and level of understanding of market conditions that characterize institutional customers and the scarcity of inter-dealer activity, quotations and transactions between a dealer and institutional customers in the same or similar securities should be as relevant an indicator of market making as quotations and transactions with other dealers. Likewise, dealers who act as market makers because of quotations and frequent transactions with institutional customers should be able to rely on contemporaneous transactions in the same or similar securities with those customers, not only with other dealers, to calculate their mark-ups. By looking only to inter-dealer transactions, the proposed definition of market maker would also have a perverse impact on dealer incentives. It would effectively penalize market makers that take the greater risks of providing liquidity to markets that are not characterized by frequent inter-dealer transactions. A dealer in a security that is subject to active inter-dealer transactions will be able to receive both the spread and a mark-up. However, a dealer of a security that has few or no inter-dealer transactions (and therefore presents greater risk to the dealer) will be presumptively limited under the Proposed Guidance to a mark-up over its contemporaneous cost.

Without the ability to benefit from the bid-ask spread as compensation for undertaking the greater risk of making a market in an illiquid security that lacks other dealer participation, a dealer could be significantly deterred from committing capital to providing liquidity to customers in these markets.

In light of these concerns, the Committees urge that the Proposed Interpretation permit a market maker to calculate its mark-up from the most recent legitimate quotation or transaction with an institutional customer in the same or similar securities. As a safeguard against any possible abuse, the Proposed Interpretation could also require that the quotation on which the mark-up is based be reasonable in relation to similar covered debt securities.

Insufficient Definition of Market Making in Illiquid Securities. The types of dealer activities that the definition recognizes as characteristic of making a market should also be expanded to better reflect the roles of dealers in illiquid markets.

Specifically, the definition of "market maker" should not set as an absolute condition the "furnishing of bona fide competitive bid and offer quotations on request" with respect to a particular security (although it certainly is highly significant evidence of market making status). Other facts and circumstances may demonstrate that a dealer is in fact a market maker, even if it is not able to show that it is furnishing bid or offer quotations in a particular security. For example, a dealer may provide quotations for some covered debt securities, and hold itself out as willing to trade in similar securities while not expressly offering quotations in those securities. Similarly, many fixed income securities are traded in markets that have limited or non-existent communication systems for furnishing quotations. A dealer in such a market may have difficulty in publicizing quotations, but nevertheless is the only source of liquidity.

A dealer in such situations may not fall within the proposed definition of "market maker" because, while it risks its capital to provide liquidity to an otherwise illiquid market, it may have difficulty in demonstrating that at any given point in time it was currently furnishing quotations in a particular security among a group of similar debt securities. In order to ensure that firms are not inadvertently penalized for providing liquidity to illiquid markets, the definition should not hinge exclusively on whether a dealer furnishes ongoing quotations in a particular debt security.

Instead, a dealer should be able to demonstrate that it is a market maker by virtue of its ongoing willingness to incur market risk to add liquidity to the market for a particular security or group of similar covered debt securities.

4. Rather than Being Presumptively Favored, Contemporaneous Cost Should Be De-emphasized as an Indicator of Prevailing Market Price.

The Proposed Interpretation states that "the best evidence of the prevailing market price is the dealer's contemporaneous cost of acquiring the securities" and that when inter-dealer transactions are not available, a dealer in covered debt securities must be prepared to provide evidence that is sufficient to overcome the presumption that contemporaneous cost provides the best measure of the prevailing market price." The Committees believe that contemporaneous cost is seriously limited and problematic as an indicator of prevailing market price, and should not be presumptively favored over other evidence of prevailing market price.

The Proposed Interpretation states that a transaction is contemporaneous if it occurs close enough in time to a later transaction that it would reasonably be expected to reflect the current market price for the security. A transaction would not be contemporaneous if it is followed by intervening changes in interest rates or other market events that reasonably would be expected to affect the market price. While this definition is helpful as far as it goes, it should include material events relating to the particular issuer as a separate factor that would be expected to affect the market price. In any event, the definition underscores the difficulties of placing heavy reliance on "contemporaneous" cost as a measure of prevailing market price. Interest fluctuations, news concerning the issuer, or other market events that may affect fixed income yields are so common that a dealer's cost would very infrequently reflect prevailing market price. This could make a cost that is truly "contemporaneous" extremely rare and hard to identify a -- peculiar for a factor that would drive a rebuttable presumption. Moreover, in unusual situations where market events have not quickly mooted a dealer's cost, that "contemporaneous" cost could understate the dealer's true cost, since it does not reflect hidden costs to the dealer, such as market exposure risk or the cost of capital charges for inventory.

The presumption attached by the Proposed Interpretation in favor of contemporaneous cost is, in our view, unjustified. As explained above, contemporaneous cost is a less reliable evidentiary tool for evaluating the fairness of mark-ups or mark-downs in the debt markets than a number of other factors. Moreover, it is both impractical and unfair to place the burden on dealers to overcome the presumption that contemporaneous cost is the best evidence of prevailing market price. The speed and volume of transactions in many covered debt securities would make it enormously burdensome, if not impractical, for a dealer to document or reconstruct its justification for relying on a basis other than contemporaneous cost for every transaction. Therefore, we believe that any of the other factors listed in the Proposed Interpretation presumptively should be better evidence than contemporaneous cost of prevailing market price. At a minimum, we believe that the Proposed Interpretation should contain a discussion explaining the limitations and marginal nature of contemporaneous cost as a guide to prevailing market price for covered debt securities.

5. Other Factors Establishing Evidence of Prevailing Market Price.

The factors other than contemporaneous cost that the Proposed Interpretation sets out are helpful. By far the most important of these factors are the three that rely on yields calculated from "similar" securities. The Proposed Interpretation correctly recognizes that transactions in fixed income securities with similar yield, credit quality and other features should be a significant factor in determining prevailing market price. The Committees believe that the criteria for analyzing when a security is "similar" are appropriate. Our only major concern with this list of factors is that it should not be taken to be exhaustive, and should not be applied in a manner that creates a major de facto record-keeping requirement. The listed factors, as well as a host of other factors that might not be foreseeable now, can influence market price. Dealers should be able to reasonably rely on any factor that affects market price. Moreover, a dealer should not be faced with an onerous burden in providing extensive documentation for the basis on which it determines its mark-up, particularly if the basis is one of those listed in the Proposed Interpretation. Rather, it should be sufficient for the dealer to reconstruct information about the factors that demonstrate a reasonable effort to set a fair price.

6. Other Observations.

"Dominated and Controlled" Securities. Footnote five of the Proposed Interpretation states that it does not address the application of mark-up policy to securities that are "dominated and controlled" by a dealer. "Domination and control" is a concept borrowed from the equity markets that has little application to debt markets. Equity securities typically are priced and trade with little relationship to other securities, while debt securities almost always trade in relation to some other set of debt securities. If the concept of domination and control has any relevance to the debt markets, the extent to which a security is dominated or controlled by a dealer should consider, as the Proposed Interoperation notes, the extent to which the security "is fungible with other or similar securities." We believe some confusion is possible here about the meaning of the term "fungible." The Committees believe that the analysis of "fungibility" should be identical with the determination laid out in the Proposed Interpretation as to when securities are "similar." Quotation Mediums. The Proposed Interpretation refers to "an inter-dealer quotation mechanism" and "validated inter-dealer quotations" as bases on which prevailing market price can be calculated. However, the Proposed Interpretation is unspecific about the types of quotation mechanisms or systems on which quotations can be posted. Quotation systems vary greatly among debt markets, particularly in light of the relative illiquidity of many of those markets. Therefore, we think it is important for examiners to apply the facts and circumstances analysis very flexibly in assessing a "quotation mechanism" or "validated quotation."

CONCLUSION

To summarize, we recommend that the final Interpretation should:

1. Rescind the statement that the appropriate mark-up or mark-down for covered debt securities is usually substantially less than 5 per cent;

2. Clarify that dealers need not sell to retail customers in order to obtain the benefit of calculating mark-ups based on contemporaneous sales prices or quotations;

3. Recognize that a dealer in a fixed income security is a market maker to the extent that it risks its capital to provide liquidity, especially if it does so in a security that is otherwise illiquid. At a minimum, such recognition involves

(i) recognizing that market-making activity can encompass activities in groups

of similar securities;

(ii) permitting market makers to rely on quotations and transactions with institutional customers in the same or similar securities to establish the basis for mark-ups or mark-downs; and

(iii) permitting a dealer to establish that it is a market maker by virtue of its ongoing willingness to incur market risk to add liquidity to the market for a particular debt security or group of similar debt securities, even if it was not furnishing single-sided or two-sided quotations in the particular debt security in question;

4 Indicate that contemporaneous cost is disfavored relative to the other listed factors in the Proposed Interpretation as evidence of prevailing market price;

5. Discuss the limitations and marginal nature of contemporaneous cost as a guide to prevailing market price;

6. Note that material events concerning an issuer are an additional factor that can affect whether a cost is contemporaneous;

7. Indicate that demonstrating reasonable reliance on the listed factors other than contemporaneous cost to establish prevailing market price does not require creating or preserving extensive documentation;

8. Clarify that "domination and control," to the extent it has any relevance to debt markets, requires consideration of the "similarity" of the security in question to other debt securities;

9. Note that the terms "quotation mechanism" and "validated quotations" should be applied flexibly to reflect the fact that systems for disseminating quotations are often less prevalent or highly developed in debt markets than in equity markets.Thank you for providing the Committees with the opportunity to comment on the Proposed Interpretation. If we can be of further assistance, please do not hesitate to contact the undersigned, or George Kramer of the SIA staff at 202/296-9410.

Sincerely,