Jonathan Katz, Secretary
Securities and Exchange Commission
450 Fifth Street NW
Washington, DC 20549
Re: Release No.34-41128; File No. SR-NASD-99-09 (extension Release No. 34-41243)
Release No. 34-41296: File No. SR-NASD-99-11
Release No. 34-41343: File No. SR-NASD-99-16
Release No. 34-41390: File No. SR-NASD-99-20
Dear Secretary Katz,
American Century Investment Management (ACIM), the investment manager for the American Century family of mutual funds1, appreciates the opportunity to comment on the NASD agency quote and related NASD proposals as detailed in the releases listed above.
American Century would like to reiterate its support of the Central Limit Order Book (CLOB) proposal submitted in the spring of 1998 (and released again aside the agency quote proposal)2. The agency quote proposal (release 34-41128) is a poor substitute for a CLOB with linkages to the Intermarket Trading System (ITS). The agency quote proposal is a confusing attempt to level the playing field for access fees that will perplex the investing public, compound quote feed distribution problems, and muddy the transparency waters even further. We believe the agency-quote proposal panders to the business needs of the wholesale marketmaking community that is trying to preserve and protect market inefficiencies directly linked to the profitability of outdated business models. Informal business discussions with market makers, institutional investors, regulators and NASD members fail to find even one supporter of the agency rule. These discussions continue to point us back to the CLOB as the proper solution. The downfall of the CLOB last year was strongly influenced by market participants who rely on payment for orderflow business practices(an economically obsolete business model with the increased transparency of a CLOB). The failure of the original CLOB proposal has hurt all market participants, and it is time to correct this mistake.
American Century supports the NASD firm quote proposal (release 34-41390) that forces market makers to move out of the way when a bid/offer greater than his/her quotation size is received and the market marker is unable to execute beyond his/her quotation size. We also support a change to the Selectnet/SOES system, but the changes outlined within the proposal submitted (release 34-41296) are again half hearted attempts while the CLOB provides solid solutions.
The CLOB vs. the Agency Quote Proposal
Routing limit orders to a CLOB, that allows for immediate cancellation, will make for tighter spreads and greater depth. This transparency, in itself, will create an environment in which it is easier for market makers to provide tight principal markets if they so choose.
A dynamic book with depth (NASD previously proposed that the CLOB depth only be available upon request - ACIM strongly opposes a CLOB that is not dynamic and continuously quoted) will make the transformation to decimalization seamless. The incrementation debate is moot when a market provides depth at multiple price levels. At least one Electronic Communication Network (ECN) has been providing this functionality for over ten years.
The CLOB provides a higher level of transparency and execution probability than existing or proposed SelecNet structures. Orders within a CLOB are executable versus quotes (whether they are "agency" or "principal"). The current market structure provides a free option to market makers when they receive the SelectNet bid/offer. Theoretically, the market maker is obligated to fill up to his/her quote size (we believe that "backing away" from the listed quote size is still rampant in the NASDAQ market) and has no further obligation. This provides an opportunity to the market maker to see the whole ticket and decide whether to execute beyond the quoted size. Market participants who set price should be rewarded with liquidity, not used as a free option by other market participants. A CLOB provides an execution facility, not an order routing facility, that rewards those who set price.
Manning rule obligations dominate many of the reasons for adopting dual market maker identifiers. These Manning obligations are much better served when these limit orders are routed to a centralized book that protects investors from "trade throughs" and safeguards orders that have price and time priority with the ability to cancel immediately.
Summarized below are seven key attributes which a CLOB must have to be successful and provide a level playing field for all market participants (these were not contained in the original CLOB proposal):3
ACIM recognizes the technological and logistical issues surrounding the parallel use of SOES and SelectNet. Order routing problems are compounded by capacity issues and the double execution risk associated with separate order routing systems. Clearly, there are problems with the current structure.
Concerns about the current structure aside, ACIM believes there are several points within the proposal that are cause for alarm. If SelectNet is to become a non-liability system, ECN's must have access to the new system for auto execution on both an outgoing and incoming basis. This is not implicit within the proposal. Such auto-execution is essential to provide equal access to all market participants. The use of ECN's is paramount to protecting the anonymity of institutional investors and in turn protecting the shareholders they represent. The order size limit of 9,900 shares within the new system is arbitrary. Will market participants be forced to route lots of 9,900 shares in rapid succession in order to be first in line? If we choose to use the SelectNet routing system, assuming SelectNet becomes a non-liability system, for larger orders, we have absolutely no recourse when these orders go unfilled. We agree with the shortening of response time from 17 seconds to 5 seconds, but we wonder why 5 seconds? Would not an immediate response serve the needs of our markets more effectively, not to mention the system capacity issues that surround multiple messages per trade.
All of the above concerns are addressed by the adoption of a CLOB with price time priority and universal access, and in many ways the adoption of a successful CLOB model will eventually make SelectNet obsolete.
During the past year, NASDAQ has been plagued by system capacity issues, would doubling the number of market maker identifiers solve this? Hardly, it is much more likely to exacerbate these capacity issues.
American Century strongly endorses a complete rejection of the recent NASD potpourri of proposals and supports the 1998 CLOB proposal with the prescribed amendments. Even the opponents of the CLOB concede that it is inevitable and they are just "buying time" to protect business models that are not economically supported in a virtual world. Certainly, there are a number of concerns within the '98 CLOB proposal, but these are addressed here and elsewhere.4 If these concerns can be addressed, and a book that provides transparency, price-time priority, depth of book, universal access via multiple entry ports and access fees that encourage quotes and liquidity U.S. equity markets will regain their leading edge and move one giant step beyond the legacy exchange architecture we are now burdened with. We request the S.E.C. quickly approve the NASD CLOB.
American Century appreciates the opportunity to comment on the proposal and would be happy to discuss any specific issues further.
Manager, Equity Trading
American Century Investment Mgt.
Century was established in 1958 and focused its business on developing
direct contact with investors through no load mutual funds. Today,
American Century manages over $90 billion in retail oriented mutual funds
with over two million shareholders.
2 Letter from American Century Investment Management to Jonathan Katz, Secretary, SEC, from Mike Cormack, May 4, 1998.
3 Letter from American Century Investment Management to Jonathan Katz, Secretary, SEC, from Mike Cormack, May 4, 1998.
4 Letter from American Century Investment Management to Jonathan Katz, Secretary, SEC, from Mike Cormack, May 4, 1998