U.S. Securities & Exchange Commission
SEC Seal
Home | Previous Page
U.S. Securities and Exchange Commission

Letter Regarding Market Data Advisory Committee Report

Peter Quick

August 27, 2001

Joel Seligman, Esq.
Dean and Ethan A. H. Shipley Professor
Washington University School of Law
1 Brookings Drive
Campus Box 1120
St. Louis, MO 63130

Re: Market Data Advisory Committee Report

Dear Mr. Seligman:

On behalf of the American Stock Exchange ("Amex" or the "Exchange"), I wanted to take this opportunity to express in writing our views regarding certain aspects of the Report of the Advisory Committee on Market Information: A Blueprint For Responsible Change (the "Report"). In particular, the Amex takes issue with the majority view expressed in the Report recommending multiple competing consolidators. While we appreciate the superficial appeal of enhanced competition in this area, we believe that in reality there will be no real competition. Moreover, we believe a move to multiple consolidators will inexorably force the Securities & Exchange Commission ("SEC" or the "Commission") to become involved in resolving rate disputes in a manner that no one appears to desire. The rest of this letter explores these thoughts in more detail.

Competition is at the heart of the national market system and the Amex endorses it. There are, however, certain types of facilities that should be exclusive because they call for centralized operations that can best be performed by a single entity. Securities clearing operations are a good example of this type of exclusive utility. Although there were five different equity clearing systems at the time the 1975 Securities Acts Amendments were adopted, today there is only one. The industry simply felt that potential competition was outweighed by the additional costs associated with operating multiple clearing organizations. We believe a central processor operates in a similar manner.

Under the existing market structure, there are two kinds of exclusive securities information processors: entities like SIAC and Nasdaq that actually process trade data, and entities like the CTA and OPRA that administer, collect and distribute market data fees. Leaving Nasdaq aside for the moment because it serves several simultaneous roles as securities information vendor, marketplace and central processor, the other central processor, SIAC, performs its service at cost. We find it hard to believe that a for-profit entity could perform the services SIAC currently performs, with the same level of service, for less money. In fact the Report itself recognizes that few vendors would actually want to compete with SIAC.

The other exclusive SIPs, the CTA and OPRA, perform a number of roles. First, they administer the market data contracts between the exchanges and end-users. Secondly, they collect revenues and distribute those revenues to the exchange participants in accordance with agreed upon formulas. Finally, they act on behalf of their participants to set a single set of fees for market data. If the majority believes that these consortia should be broken up to create competition, we believe this conclusion is erroneous.

Looking at the different functions performed by the consortia, it is hard to envision how competition will work to the benefit of the industry or public. The Report cites three potential benefits: potential innovation, reduced administrative costs and possibly enhanced competition among market centers. The Amex does not believe there is any reliable evidence that any of these supposed benefits will in fact occur. With respect to administrative functions such as contract administration, we cannot imagine how it will be more efficient to have eight market centers administering their own contracts when we currently have a SIP administrating one contract for all of them. Second, with respect to potential systems innovation, it is hard to imagine how the industry or public will be served by having multiple vendors performing exactly the same service that SIAC performs today at cost. It would appear that there would simply be redundant systems and redundant expenses. Finally, we believe the consortia, while not perfect, are an important means of bringing competitors to the bargaining table. Dismantling them will simply remove one avenue for resolving their differences with respect to these very important national market system issues.

What the Report does not say is that the real affect of breaking up the consortia would be to eliminate the current unified fee structure and allocation formulas. Thus, if the consortia were dismantled, each market would set its own fees. Presumably, the majority believes that these fees would reflect the value of each market's data to end-users. Thus, the proposal for multiple competing processors is not really to create competition in the processing business, but rather a fairly transparent way of altering market data fees and the allocation of those fees.

Market data fees are currently set by the consortia and charged to end-users. Securities information vendors do not buy data from the consortia; they are merely responsible for monitoring the number of devices installed by end-users and forwarding that information to the consortia. If the consortia are disbanded, presumably each market will attempt to assess fees on end-users that are the same or higher than the fees the exchanges receive today from the consortia. If the Vendor Display Rule remains in effect, each vendor will continue to be required to acquire data from each market and disseminate it to end-users. The end-users will have no choice whether to receive the data since vendors have no choice whether to disseminate it. Thus, end-users will have no negotiating leverage to restrain the enthusiasm of the exchanges to charge exorbitant fees. The Vendor Display Rule will give smaller exchanges the market power that is identical to the larger markets.

The majority believes that brokerage firms that send order flow to the exchanges will force the exchanges to charge appropriate fees. While this might be true, it could be equally true that it will not, in which case the SEC will be forced to be the arbiter of ongoing and endless fee disputes. Moreover, separately negotiated fees could change the fundamental economics of exchanges in unpredictable and inappropriate ways. For example, separately negotiated fees could cause the regional exchanges to lose the bulk of their market data fees. As a result, competition would be fundamentally lessened not strengthened. Finally, the advantage of the current fee structure is that it is immediately responsive to market share changes that, at least at one point in time, were thought to be a proxy for the value of market data. Moving to an environment, in which each market would, on an individual basis, determine the value of its market data, would necessarily be less responsive to subtle shifts in the value of that data, however determined.

Rather than radically change the world in this way, the Amex believes it makes more sense to continue to have a unified fee structure. This will at least keep the number of disputes down to one per market data stream rather than the potential of eight different fee disputes. Concomitantly, we believe a better way to address the real concern driving the majority view (namely what is the value of market data) is to reformulate the allocation scheme to better reflect the value of that data.

At the time the allocation formula was created (at least for the CTA), the only market data available was last sale information. Today, there is both last sale data and quotation information. It would be difficult to argue that NBBO quotation information is not at least as valuable (if not more so) than last sale information. Moreover, it may be that in a decimal world, depth away from the market may be more valuable than the NBBO. As a result, we would favor a reworking of the allocation formula at least to reflect the value of quotation information. Marketplaces that provide market depth of market and add to price discovery by publishing competitive quotations should be compensated. In addition, we also support other changes to the structure of the consortia described in the Report, such as expansion of the governance structure to include industry and public representatives.

We have taken the time to put our thoughts in writing because we feel strongly that the majority view is misguided. If, after reviewing these thoughts, the majority is still in favor of dismantling the consortia, we respectfully request that our views be appended to the report as minority views.

Yours truly,

Peter Quick



Modified: 09/18/2001