October 12, 2000
Mr. Jonathan G. Katz, Secretary
U.S. Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, DC 20549
Re: SEC File No. SR-NASD-99-53
Dear Mr. Katz:
Renaissance Technologies Corp. ("Renaissance") is a registered investment adviser engaged, among other things, in the mathematically based trading of equity securities on behalf of the private investment funds it manages. As a significant participant in the equities markets, Renaissance has studied with great interest the NASD's proposal for the creation of a new order display and matching facility ("Super Montage"). In light of the numerous detailed comment letters that have already been submitted with respect to the SuperMontage proposal, Renaissance hereby seeks to offer only a few general observations from the viewpoint of an end user which conducts a substantial volume of trading in NASDAQ securities through ECNs. We also address the important issue of costs, which has been largely ignored by the comment letters previously submitted.
In today's markets, large institutional investors such as Renaissance provide significant liquidity through the volume and frequency of their trading, thereby narrowing spreads, increasing the likelihood that limit orders will be executed, and generally improving market efficiency. This institutional liquidity has grown dramatically in recent years as trading costs have declined. For such institutional investors cost is of paramount importance. SuperMontage as currently proposed will be an expensive proposition for high volume, active traders and ultimately the individual investors whose money pays these trading costs. Moreover, as has already been well articulated in numerous comment letters heretofore submitted, we believe that SuperMontage is monopolistic and will chill technological innovations and eliminate the healthy competition among ECNs for business. What may have been overlooked is the extent to which the competition from and among ECNs is directly responsible for reducing costs, making the U. S. markets the world's most efficient markets.
The NASD does not have an impressive track record in providing low-cost, efficient trading systems. Consider SelectNet, which SuperMontage would replace. SelectNet charges a $1500 monthly connection fee (its SDP fee), plus a dollar for each order routed over SelectNet to a market maker or an ECN. The Island ECN, on the other hand, charges on average half as much, with no monthly fee, to take in two customer orders, cross them, and clear the trades, where clearing makes up most of Island's costs. How can an ECN do so much more for so much less? The answer is that Island has competitors while SelectNet has a unique relationship with the NASD, which is the only explanation for charging higher fees and providing less service. Although the NASD has not yet committed to a fee schedule for SuperMontage or SuperSOES, it is hardly encouraging that these fees are anticipated to be no greater than those charged for SelectNet. It is axiomatic that monopoly creates no incentive to innovate or control costs.
Renaissance is particularly concerned because SuperMontage ties all participants into one network, which therefore makes all NASDAQ trading dependent on the technological capability of one system. We question whether NASDAQ's SuperMontage systems will be technologically able to withstand the pressures of demand and volume and worry about the potential for systems failures which would effectively shut down trading in NASDAQ securities. Large institutional investors prefer to diversify their risks, not concentrate them. We believe that the real risk of potentially catastrophic loss far outweighs the perceived benefits of putting all the eggs in the SuperMontage basket.
Furthermore, we believe that the issues the NASD is trying to address with SuperMontage would in large part disappear if market makers were compelled to respond to orders in the same time frame as ECNs. Today, ECNs must respond immediately while market makers have 90 seconds to respond, giving them the opportunity to "back away". We doubt that the need for a centralized order routing facility would exist if market makers were truly competing with ECNs on a level playing field.
Finally, Renaissance is troubled by the conflicts of interest inherent in SuperMontage. The NASD is a regulatory authority and it undermines the perception of neutrality and impartiality implicit in that role if the NASD exploits that very role and its privity to market data for commercial and competitive purposes. Moreover, because of the NASD's imprimatur, there is a coercive element to participation in SuperMontage. The perception that trading on the NASD's ECN will discharge best execution requirements Q.E.D. is a powerful tool in compelling large market players to lend their liquidity to SuperMontage. We think this does nothing to promote the integrity of U.S. markets and find the whole idea unsavory.
We believe that SuperMontage, even in its latest iteration, is based on the wrong premises and should be rejected by the SEC.
Yours very truly,