New York Stock Exchange, Inc.

Comments to
Securities and Exchange Commission Concept Release on
"Regulation of Market Information Fees and Revenues"
(Release No. 34-42208; File No. S7-28-99)


Cover Letter on Cost-Based Regulation

PHB Hagler Bailly, Inc.
1776 Eye St, NW
Washington, DC 20006-3700

PHB Hagler Bailly, Inc
1776 Eye Street, N.W.
Washington, DC 20006-3700
Tel 202-828-8760
Fax 202-429-1837
Alan Kolnik
Senior Vice President

March 30, 2000

Mr. Robert G. Britz
Group Executive Vice President
Market Data
New York Stock Exchange
11 Wall Street
New York NY 10005

Re: SEC's Request for Comments to its Concept Release No. 34-42208 (File No. S7-28-99)

Dear Mr. Britz:

At your request, we have reviewed the above Concept Release in detail.

PHB Hagler Bailly is a well established consulting firm based in the Washington D.C. area with more than twenty years of experience advising clients on matters of regulation across almost all the regulated industries. We have used our experience and expertise to examine the Concept Release, and are pleased to attach the findings of one of our economic experts, Dr. Oliver Grawe, Ph.D., and one of our regulatory experts, Ms. Bruce McConihe. I have contributed my own knowledge of the industry from both the vendor side and my experience and knowledge of CTA/CQ and OPRA activities.

Dr. Grawe and Ms. McConihe have extensive experience in regulatory matters and have participated in many of the major regulatory and anti-trust actions of the last twenty years. We believe that the SEC will find their views on the nature of regulation, the need for regulation, and the consequences of regulation applied to the matter of pricing of market data insightful and worthy of thorough consideration.

If we were to summarize our findings in one paragraph, it would be to say that the SEC, through a process of thoughtful application of the Congressional mandate for the regulation of the Securities Industry (particularly following the 1975 amendments to the Securities Exchange Act of 1934), has arrived at the type of beneficial regulation of market data prices which is currently the goal of many other regulatory bodies who are moving away from cost-based ratemaking.

It is extremely surprising, in light of the accumulated experience of regulatory bodies across the regulated industries (which we have thoroughly documented in our paper), to observe the SEC proposing to move against the trend away from cost-based regulation. Implementing the approach suggested in the Concept Release would move the securities industry to cost-based ratemaking of prices for market data - a regulatory approach which has been superseded in other regulated industries by regulation far more akin to the SEC's current approach. This can only be regarded as retrogressive.

In the first place, the case has not been made that there is a need to change the current consensus-based method used by the industry's exchanges and customers and to increase regulatory oversight of market data prices. The SEC itself has acknowledged the success of the present system in making market data widely available1. Prices for market data, as shown in the tables attached to the Concept Release, have declined. Furthermore, examination of financial services sites readily available to all consumers over the Internet reveals that many services make real-time data available at no cost (free) to the public. The pricing structure that allows professionals and non-professionals alike unlimited quotes for a fixed monthly fee2 means that the marginal cost of an extra quote is zero. Of course, there is no charge at any level for delayed market data. It is hard to imagine additional regulation improving on this situation, particularly in view of the fact that market data fees represented less than one quarter of one percent of the securities industry's costs in 1998.

We stress at several points in our attached papers that the objective of rate regulation in other industries has never been solely to minimize the rates to consumers, since very low rates may affect the desirability for the service provider(s) of staying in business or the quality of the product being provided. A regulatory approach that would reduce revenues from market data prices without considering the financial situation of the exchanges could result in the business failure of one or more exchanges3 or attempts to make good the shortfall by, for example, raising transaction fees or reducing the cost of ensuring quality (e.g., by reducing the level of surveillance and/or self-regulation an exchange is prepared to undertake).

We demonstrate in our paper "The Economic Perspective on Regulation of Market Data" that there is sufficient competition in the securities industry to obviate the need for further regulatory intervention in the matter of market data prices. Market forces have in fact provided the discipline that any regulator would expect and desire in the area of market data services and fees. However, if the conditions existed to justify regulation (and this is not the case), requiring a firm to set individual prices at marginal cost guarantees insolvency. Economic theory and practice make it quite clear that regulation designed to serve investors' well-being would not limit rates for any regulated service to the marginal cost of making the service available. Marginal cost pricing only works well when the very conditions leading to a need to regulate are not present. Hence, the "marginal cost" approach cannot be the appropriate regulatory benchmark.

When substantial economies of scale and scope dictate that competition fails, then a regulator concerned with consumer welfare, but recognizing that the supplier must be viable, would set prices that depart from marginal cost. The departures may be substantial when demand is inelastic. This result is not inherently unfair. When economies of scale and scope exist there are significant "shared" costs and a benign regulator would consider the full cost of creating all services offered, and price individual services based on the elasticity of demand for those services.

Modern regulatory theory realizes that a more effective regulatory model for active regulation of prices than cost-based pricing is a model that tries to mimic competition. This is the theory of Ramsey4 pricing, which accounts for consumer welfare while meeting the need to keep the supplier viable. However, Ramsey pricing is complex. Ramsey prices cannot be deduced from cost data alone. Rather, prices for any one service are arrived at by assessing the elasticity of demand for that service. Those services with a lower elasticity of demand are priced to recover a greater portion of shared costs.

The SEC has proposed a flexible cost-of-service approach to rate regulation which it believes will make the process less onerous. This is overly optimistic. As a practical matter, rate regulation cannot be implemented without first undertaking the rigorous cost analysis that the SEC wishes to avoid (recognizing its complexity). Further, the revenue distribution scheme, which apparently would not be based on each SRO's costs, appears impossible to implement. Under this scheme, some SROs would under-recover, and some would over-recover, their costs. A revenue model that requires that aggregate revenues be based on aggregate costs requires an allocation scheme that allocates revenues to each exchange based on its individual costs.

An unexpected consequence of cost-based regulation would be that the SEC would add the cost of compliance to the ratebase due to additional monitoring and audit functions it would require SROs to perform. Such requirements are onerous and costly, especially given that the current system is working satisfactorily. Under the proposed cost-of-service methodology, prices would be set using a rate of return on the cost base that the SEC would determine, including these new costs. By adding to the cost of creating market data, the SEC might unexpectedly raise the prices to consumers.

To conclude this introduction, one should note that rate of return regulation has proven to be both extremely litigious and extremely costly. There are numerous examples of the contentious nature of this form of regulation in our paper "Issues Surrounding Cost-Based Regulation of Market Data Prices" (the "Issues" paper). Closer to home, the Instinet proceeding5 should provide ample warning of what is likely to come. As far as costs are concerned, regulation increases the burden on both the regulator and the regulated parties. In fact, one authority has estimated that for every dollar spent by the federal government in regulating industries, the regulated firms and other government agencies spend forty dollars.6

The structure of our findings, in the documents that are attached, is as follows:

Issues Surrounding Cost-Based Regulation of Market Data Prices ("Issues" Paper)

Market Data - the Original Conception

The Senate Report ("S. 249") to the 1975 amendments of the Securities Exchange Act of 1934 states:

The Commission was directed [in the Senate bill] to remove existing burdens on competition and to refrain from imposing, or permitting to be imposed, any new regulatory burden `not necessary or appropriate in furtherance of the purposes' of the Exchange Act. (Page 94)

If the Commission adopts a cost-of-service regulatory model, as the Concept Release proposes, rather than refraining from adding regulatory burdens, there will be significant additional burdens created for the SROs in terms of functionalizing costs, designing rates and allocating market data revenues. There will be winners and losers depending upon the specific model adopted. In addition, there will be significant new burdens placed on the SEC itself as it attempts to oversee the industry using a cost-based model. We describe these unexpected effects in detail in our report.

Congress did not envision that the SEC would pervasively regulate market data. S. 249 clearly states that:

This is not to suggest that under S. 249 the SEC would have either the responsibility or the power to operate as an `economic czar' for the development of the facilities of a national market system must come from private interests and will depend on the vigor of competition within the securities industry as broadly defined.

It is clear from S. 249 that Congress envisioned competitive market forces within the industry to encourage the development of centralized dissemination of market data. This is just what has happened since 1975. We discuss the nature of this competition and its positive effects in our economic analysis.

Demands on the SEC if it moves to implement cost-based ratemaking

In addition to the potentially harmful effects on the regulated entities of adding to the burdens which Congress specifically requested be reduced (see above), the SEC itself would be subject to a new level of demands and burdens that would flow from the complexity of overseeing a cost-based ratemaking scheme.

In order to illustrate the complexity that the SEC will face, we have laid out the steps that the SEC will need to undertake in order to implement cost-based regulation. These are numerous and onerous. We refer you to our "Issues" paper, for a full reading. In summary, the SEC will require many more staff, incur far greater expense, and be subjected to many more rate hearings and, potentially, far more litigation than has been the case until now.

In addition to the step-by-step description of the process and activities to be implemented that is provided in our paper, it is worth noting that there would be unexpected consequences of additional regulation that would act to increase the burden on the SEC (and the SROs). For example, under the Paperwork Reduction Act, agencies must estimate the paperwork burden involved with their actions and have that approved by the Office of Management and Budget.

The change in regulation proposed under the Concept Release would result in additional burdens in the form of additional accounting work, audits, rate hearings, cost studies, and other activities associated with cost-of-service regulation. The SEC will need to estimate the magnitude of these burdens in terms of time and financial resources so that parties potentially subject to the burdens have opportunity to comment and so that the estimates can be incorporated into the economic (cost/benefit) analysis that is to be performed by the regulatory agency.

Distortions introduced by rate regulation

In our "Issues" paper, we examine the basis for regulation, its history, and consequences. We lay out the process of cost-based regulation, as it has been implemented, and the consequences, as they have been experienced. In other words: "If you wish to regulate the price of market data using cost-based regulation, here is what must be done, and these are the results you can expect".

We start with the point of view that regulation is an economic, legislative and legal concept. The legislature usually determines which industries will be regulated. These decisions are usually based on consideration of two classes of issues:

However, the policy adopted must conform to existing legal concepts and procedures. This is the most critical aspect lacking in the SEC Concept Release from the regulatory standpoint. Regulation is not created without precedent, without consideration of appropriate rates of return, and without reference to the ability of the regulator to withstand challenges to their proposed regulation and rates. The regulatory scheme the SEC devises must be equitable and robust enough to withstand challenge from both providers of market data (current exchanges and potential new entrants) and customers - while increasing or at least preserving the industry's health and the customer's well-being.

The Concept Release proposes a cost-based regulatory method that it refers to as a flexible, cost-based approach. There is, in fact, no such methodology as "flexible costing". Regardless of the label one might try to give it, cost-based regulation could never be implemented without first undertaking a thorough, rigorous cost analysis of all the exchanges in order to set up the chart of accounts which would be used to decide what costs to include and what costs to exclude. There is nothing flexible about this process - it must be conducted in a rigorous, disciplined and thorough manner. The very "flexibility" (if it could be achieved at all) that the SEC suggests would reduce the regulator's burden would, instead, actually open the door to endless disputes as different parties attempt to leverage the lack of rigor and clarity implicit in such a concept to sway ratemaking decisions in their favor.

Based on the experience of other industries that have employed cost-based ratemaking, it is likely that there will be an increased burden on both the industry and the SEC. The burden could be enormous, especially bearing in mind the need to oversee the current quite large set of exchanges and the future entrants. The regulations must cover all industry players, not any one or a particular subset, except for relief given to special cases - such as very small players, for example. We detail the histories of nine industries to demonstrate this issue and to show that regulators have moved away from cost-based regulation. These histories detail the kinds of onerous activities and costs the SEC would have to undertake in order to accomplish its goal of cost-based ratemaking. Moreover, since there is ample precedent that the costs of such burdens may be recovered in the rates the regulator establishes, the additional oversight would have the effect of adding to the rate-base, and, therefore, the rates.

Distortions Introduced by Rate Regulation

Traditional Public Utility Regulation - The Process

Price regulation is the key element of public utility regulation. The basic standard used to assess rates is the revenue requirement standard. Regulators must allow utilities to set rates at levels that will generate aggregate revenues sufficient to recover operating expenses and allow a rate of return that permits them to replace plant that has reached the end of its life and to invest in improvements for the benefit of consumers.

This requirement creates, at a high level, a four-step process that a regulator must follow:

One should note that, since these steps must be executed in sequence to create the regulatory price structure, there appears to be no way to "save a step", so to speak, by using some kind of "flexible" methodology. The implied lack of rigor would open any findings or regulations to endless challenges which could themselves only be refuted by attempting to rigorously justify their basis - which would require doing the necessary work of cost analysis and allocation. There is circularity here which cannot be overcome.

Before any cost analysis can be performed, the regulators must have in place an accepted uniform system of accounts so that the utility's financial statements reflect commonly accepted rules for allocating expenses and revenues to functional categories.

An important element of a uniform system of accounts is the allocation of costs (administrative and property) that is used in common (joint) to serve different customer classes, provide different services, or supply both interstate and intrastate services. Any cost allocation method involves elements of arbitrariness and the results obtained may vary significantly depending on the allocation method chosen. Thus, the regulator must ascertain which costs, in its opinion, will be allowed for purposes of calculating the rate-base.

Having established a uniform set of accounts, the regulator will be able to turn to the most important component of a public utility's rate level - allocating operating expenses and capital outlays to be included in the ratebase for market data. Allocations of operating costs and capital outlays that responded immediately to changes in the business environment would require regulators to participate in the utility's moment-to-moment, transaction-by-transaction decision-making process in connection with every aspect of a utility's business. This is not possible7. Therefore, the SEC will have to determine general rules to guide the allocations of the various costs allowed by the uniform set of accounts that will be used to determine the cost-base for market data.

The SEC will also need to determine how it will "certify" the SROs and what it will do if firms try to enter the industry as exchanges. Currently, several ECNs have filed or are considering filing for exchange status. Therefore, in addition to any rules that are in place regarding what these firms must do in order to become SROs, the SEC will need to promulgate rules that describe how they will be dealt with from the perspective of cost-based regulation. These rules would have to account for the new entrants' accounting and for changes in the rate-base of the "traditional" exchanges. For new entrants that are currently part of the NASDAQ cost-pool, their costs would need to be separated out along with their revenues. This would mean that a new cost proceeding would have to be performed in order to determine how much the new SRO's rate base should be and how much NASDAQ's rate base had changed.

The value of the asset base used to provide the service will need to be determined and a depreciation schedule will need to be calculated in order to allow capital recovery to be achieved via the rates. This is complex in the normal scheme of things. In this case, given the need to allocate the portion of the capital plant of each exchange that should be included for the purposes of setting cost-based rates for market data, the complexity will be even greater. The SEC will need to attempt to match capital recovery with capital consumption:

We discuss appropriate methods in our paper, "Issues Surrounding Cost-Based Regulation of Market Data Prices".

Although the use of depreciation is quite routine in public utility regulation, methods of handling obsolescence are not well-established. In an industry so dependent on information technology, and given the pace of change in information technology, the issue of obsolescence may pose an even larger challenge for the SEC than it has for other regulators.

The most highly controversial of the cost elements that regulators deal with is the matter of the level of profits (the rate of return) that will be allowed on the utility's investment to provide services to consumers. We examine the issues around establishing appropriate rates of return in our paper.

Firms in an industry must earn at least a "normal profit" or the firm will exit the industry because they could earn a higher return on their resources in other areas. Normal profit is a return just equal to the opportunity cost of the entrepreneur's effort, investment, and special skills. Thus, the courts have also determined that firms must be allowed to earn a "reasonable return" or else they would be considered subject to an illegal "regulatory taking." Furthermore, since a firm's profit is determined by a variety of factors (price being just one of them), a regulatory agency may have a difficult time in the determination of a normal rate-of-return.

Although the NYSE, as a mutual organization, may not be able to enter other lines of business, under sufficiently adverse conditions it (or, more likely, other SROs) could be forced to exit the business. Following the above argument, the regulator (the SEC) would have to ensure at least "normal profits" for the regulated entities in the securities business, both from the common-sense point of view that it should not do harm to the industry it is charged with regulating and from the point of view that it avoid incurring charges of "regulatory taking".

Determination of rate of return by regulators requires resolution of what rate constitutes a "fair" rate. "Fairness" implies fairness to both the regulated firms and to the consumers whose well-being is sought. Establishing a rate structure that will provide fair and efficient incentives for all parties is not a trivial undertaking. The objective of rate regulation in other industries has never been solely to minimize the rates to consumers, since very low rates may affect the desirability for the service provider(s) of staying in business or the quality of the product being provided. In addition, minimizing rates may impact the incentive to innovate, so that the exchanges will not be able to improve efficiency and thus lower rates over time.

It should be noted that representations to the SEC over this issue of fairness have been a principal source of the SEC's interest in the matter of market data pricing. It is reasonable to suspect that the acrimonious and vigorous advocacy of different standards of "fairness" will occupy the SEC considerably more than it already has if it implements cost-based ratemaking, as has been the case in other industries where cost-based ratemaking has been employed.

Problems With Cost-Based Regulation

In order to investigate the consequences of cost-based regulation, we examine the problems with this method of regulation that have occurred in the past.

Fundamentally, the problems arise because regulators must use quasi-scientific tools that involve judgment and decisions to attempt to mimic a competitive market environment. These judgments invariably influence the market outcomes and create distortions in the marketplace. In addition, the process is backward-looking in that it develops prices based on historic information and market structures. In an industry experiencing rapid change due to technology, it is not possible for regulators to anticipate future market structures (in terms of supply and demand) in setting prices. Prices based on cost-of-service will never reflect current market conditions, but rather the regulators' best approximation of what might occur.

Based on this point-of-view, it is hard to find an industry less amenable to cost-based regulation than the securities industry because the pace of change in technology, competitive landscape, investor preferences, and so on, is so rapid.

The Concept Release provides an interpretation of Bonbright's approach to regulation. A deeper reading reveals that, in fact, Bonbright would concur with pricing market data at levels that account for differences in consumer demand. He acknowledges that the failure of the cost principle is that it does not give direct weight to value of service to the consumers as distinct from the cost of production to the producers. In a different way, this repeats the economic argument in favor of establishing demand elasticity as the method for price determination.

The example of the attempts by the Federal Power Commission (FPC) to administer a regulatory scheme for natural gas describes the issues that can arise from cost-based regulation. The FPC ultimately rejected this regulatory approach, stating: "If our present staff were immediately tripled, and if all new employees would be as competent as those we now have, we would not reach a current status in our independent producer rate work until 2043 AD-eighty-two and one-half years from now." However, the price-setting strategy it then implemented had additional unexpected consequences, leading ultimately to production shortages of natural gas. The difficulties encountered should be considered a cautionary tale by the SEC as it plans its regulatory strategy.

Lengthy, costly and litigious proceedings seem to be an unavoidable by-product of regulatory proceedings. We cite examples of the length of time it has taken in other industries to settle ratemaking cases, and point out the similar experience in the Instinet case in this industry.

Finally, we point out that partial regulation (regulation of only some of the services provided in an industry) creates yet another level of complexity.

This problem arises with the Concept Release proposal. Total costs will be determined for all SRO activities and a portion of the costs will be allocated to the provision of market data services. Cost-based regulation is not conducive to partial rate regulation. Whatever rate structure is adopted for market data services will importantly affect the revenue requirement of other services provided by the SROs.

Partial regulation of market data services will inevitably extend into examination and costs of the other services provided by the SROs in order to determine "appropriate" allocations. For example, a customer may contend that too much of shared costs are allocated to market data, while a provider may claim that too little is. The SEC will have to resolve the matter, and will therefore need to examine all costs in order to perform the allocation exercise. Thus, a quick regulatory solution of moving towards cost-based rates in order to limit profits and potential discrimination may evolve into larger problems concerning cost recovery in the other services provided by SROs.

Furthermore, any cost-based regulatory scheme implicitly incents the regulated entities to raise costs through inefficiencies, or at the very least, not to eliminate inefficiencies in order to reap larger absolute rewards based on their rate-of-return. The unexpected consequence of this may well be that over time market data fees rise (with regulatory approval!) - as opposed to the long history of declines that can be observed under the current regulatory approach.

Problems raised by the "Flexible Cost-of-Service" Concept

We have examined the Concept Release's proposed flexible cost-of-service ("Flexi-COS") in detail in light of the above. It is possible to foresee a number of difficulties, explained in detail in our "Issues" paper. Specifically:

The Passing of Regulation - Deregulation

In order to demonstrate the complexity of cost-based ratemaking and to substantiate the significant trend away from this form of regulation, we provide capsule histories of regulation in major industries in which regulation has had a significant role. The industries are:

Furthermore, we list the major economic deregulatory initiatives from 1971 - 1989 and forms of deregulation that have been implemented.

The Economic Perspective

In addition to reviewing the issues that cost-based regulation would raise from a regulatory standpoint, we also reviewed the economic issues that would affect the cost-based regulation of market data pricees.

Is There a Need to Regulate?

Economic price and access regulation has historically been reserved for cases where technology has resulted in economies of scale and scope that are so significant and enduring as to result in a "natural monopoly" or an "essential facility."8 It is far from obvious that primary markets display these features. Competition for order flow (transactions) and for listings is significant. Moreover, changes in technology, including changes permitting round-the-clock trading in primary-market listed equities - trading that occurs even when the listing exchange is closed - suggests that price discovery activity need not reside or occur in a single market.

The Concept Release frequently refers to its goal in terms of the impact on "final consumers", the retail investors.9 The SEC states: "broad access to real-time market data should be an affordable option for most retail investors, as it long has been for professional investors."10 The SEC's focus on retail investors, as opposed to professional investors whom the SEC apparently concludes have adequate access and currently pay "affordable" rates, includes the premise that market data fees for retail investors "must not be unreasonably discriminatory when compared with the fees charged to professional users.11

The public information available from even a casual Internet search indicates that free or low-cost real-time market data is very widely available to retail investors for exchanges located in North America.12 The real-time data available from many sources, including on-line brokers or even information services such as, is free.13 Direct economic price regulation is unlikely to improve upon free real time data insofar as retail investors are concerned. This does not square well with an obvious and compelling need to regulate, as the potential consumer gains from doing so are miniscule at best.

What is The Nature of Regulation Designed to Serve Investors' Well-Being?

It is important to stress one critical point that economic theory and practice make very clear. Regulation designed to serve investors' well-being would not limit rates for any regulated service to the marginal (incremental, direct) cost of making the service available.

This issue strikes at the heart of any attempt to price market data using a marginal cost-of-service allocation method. There are at least two principal reasons for this position, well established in economic and regulatory theory and practice:

  1. A regulator interested in maximizing consumer welfare but obliged to mediate between contending consumer groups arguing over how to allocate shared costs would allocate the bulk of the shared cost to the service(s) with the least elastic demand. And the service with the least elastic demand may well be market data.

  2. The regulated firm is viable only when the sum of the rates charged for all the firm's services covers the incremental cost of providing all of them together. That sum is the firm's total cost including the shared cost.14

Hence, the fees for each service must differ from incremental or direct cost. In a very real sense, price regulation of a multiple-service firm runs the risk of making the regulator the vehicle for negotiating how shared costs will be allocated between different groups of buyers, who may or may not be competitors of one another in their own output markets.

This effect is already observed in submissions by industry participants to the Commission, in which they attempt to sway the Commission towards regulatory actions that would favor their competitive model (inevitably, at the expense of their competitors, who would be correspondingly disadvantaged). Thus, when the Commission is called upon to rule on, say, new pricing mechanisms for market data, it will have to consider the impact of its actions not only on the purported beneficiaries (retail investors, say) but the new competitive situation that will arise in the industry. And upon ruling one way or another, it can expect to be subjected to new demands by disaffected competitors attempting to redress the damage they perceive has been done to their cause.

The best regulation is the least regulation. In fact, the SEC has achieved a consensus-based ratemaking approach that has worked well, as the SEC staff point out in the concept release. As mentioned above, the most effective regulatory model is one that tries to mimic competition. This is the theory of Ramsey pricing, which is intended to achieve similar results, but is considerably more burdensome than the SEC's current approach. Ramsey pricing begins with consumer welfare in mind, and accounts for consumer welfare while meeting the need to keep the supplier viable. Ramsey prices cannot be deduced from cost data alone. In fact, prices for any one service are arrived at by assessing the elasticity of demand for that service, not by a simple accounting allocation of cost. We discuss this approach in detail in our paper and briefly below. Note that implementation of Ramsey pricing requires the regulator to perform difficult analytical tasks of cost and demand elasticity examination.

The correct rule prices must meet is that the price for each service and for each bundle of services including the bundle that consists of all services must cover the incremental cost of making that bundle available. Thus, aggregate revenues (earned by summing the revenues from each service) must cover aggregate costs (incurred by providing all services). However, the price for any particular service should not be arrived at by trying to estimate the costs incurred just for that service because of the difficulty of assigning shared costs correctly to multiple services. Prices for each service are arrived at by assessing the elasticity of demand for that service, not by an accounting allocation of cost, which is always, in the final analysis, arbitrary. (Typically, accounting allocation schemes use rules such as equipment or staff share of floor-space, for example, to determine overhead allocations to particular goods or services. What fraction of building security costs, for example, should NYSE allocate to market data as opposed, say, to market operations costs?)

A key point emerging from this well-accepted approach is that consumer-welfare maximizing prices set by a benign regulator would invariably depart from marginal (incremental or direct) costs and possibly by a significant amount. For example, if transaction demand is much more elastic for NYSE than is the demand for NYSE's market data, a benign regulator would impose most of the shared costs associated with running the NYSE equities' auctions on market data fees. Data fees might rise if this proves to be the case.

However, it is important to bear in mind the unintended consequences that can accompany even the most-well intentioned regulation. For example, there are social costs such as slowing down price adjustments due to the need for a hearing every time a price change is requested (by a vendor or a customer, of whom there are multitudes). There is the reduction of innovation seen in regulated industries such as the telecommunication industry before the introduction of competition in 1984 and, particularly, in 1996. We provide examples of these matters in detail in our "Issues" paper. Thus, rather than instituting a new regulatory approach, the SEC might wish to consider whether the current consensus-based ratemaking it has already achieved might not have already reached a better outcome.

What Will be the Effect of Additional Regulation on Competition in the Exchange Environment?

Investors do not need, under current SEC regulations, to access any specific exchange - or any exchange - to consummate trades once they have market data. Competition for transactions - for order flow - is substantial and places a cap on the revenue stream any exchange can earn from transaction fees.15 Maladroit price regulation on market data fees - limiting market data fee income beyond market-based limits - could well have unfortunate consequences, because such regulation could lead to unwanted and unexpected changes in other fees, including transaction fees. However, as mentioned above, competition may limit the degree to which exchanges can raise transaction fees to compensate for lost market data revenue. In addition, as transaction fees rise, to the degree that they can, a particular exchange may find that transactions migrate to exchanges that are less costly to the investor. An exchange in this situation would lose both transaction revenue and market data revenue - clearly, two important sources of funds.

Thus, the SEC should examine the potential for the failure of one or more markets as transaction fees rise and transactions migrate if market data revenues are forced down. Moreover, Tables 2 and 3 in our "Economics" paper illustrate the dependency of many of the smaller exchanges on market data fees. (NYSE is not nearly as dependent on these fees). Smaller exchanges could easily become non-viable if this source of revenue were reduced.

*     *     *    *    *

To recapitulate, the key points in our findings are:

We are pleased to submit these finding to you in support of your submission to the SEC. We suggest that you urge the SEC to reconsider its approach in this matter.


Alan Kolnik

Senior Vice President

1 See, e.g., the Concept Release: "All participants in the U.S. markets have access to a consolidated, real-time stream of market information for any of the thousands of equity securities and options that are actively traded. The information for each security is "consolidated" in that it is continually collected from the various market centers that trade the security and then disseminated in a single stream of information." (Page 4) "Under this regulatory framework, the SROs have developed and funded the systems that have been so successful in disseminating a highly-reliable, real-time stream of consolidated market information through out the United States and the world." (Page 4) See also, "The NYSE historically has operated and regulated one of the largest and most prestigious markets in the world and has, as well, taken a leading role in the regulation of its members, which include most of the largest broker-dealers." (Page 23)
2 The fee per terminal for professional investors is fixed; however, the firm at which the professional works pays a per-terminal rate that results in an aggregate fee that depends on the number of terminals, up to a cap of $500,000 per month. For any given number of terminals, the monthly fee is fixed independent of the number of quotes retrieved. For a fee of $1.00 per month or less, non-professionals may access as many quotes as they wish. Thus, in either case, the marginal cost of data is zero.
3 See Tables 2 and 3 and analysis in our attached paper, "The Economic Perspective on Regulation of Market Data Prices".
4 Ramsey pricing derives its name from Frank P. Ramsey, "A Contribution to the Theory of Taxation," Economic Journal, Vol. 37 (1927).
5 NASD filed proposed fees for Instinet on June 17, 1983. Instinet protested the fees on July 15, 1983 and insisted that NASDAQ's proposed fees must be cost-based. The matter was not settled until 1990 after numerous hearings, proposals, counter-proposals and litigation (National Assoc. of Securities Dealers, Inc. v. SEC). The Concept Release points out that the SEC noted that, "in reviewing the fairness and reasonableness of the proposal, the Commission finds it significant that the proposed fee of $50 is the result of negotiations among the concerned parties [emphasis added] after protracted proceedings." (See Concept Release, page 20 -21 for a full description).
6 Murray L. Weidenbaum, Business, Government, and the Public, 2nd ed. (Englewood Cliffs, NJ: 1981).
7 Claire Wilcox, Public Policies Towards Business, 4th ed. (Homewood, Ill.: Richard D, Irwin, 1971), p. 477-8:

"The regulated industry comes, in the end, to have two masters: its own management and the regulatory agency. Essential functions of management are duplicated. Managerial decisions are reviewed. Where the regulatory agency finds them to be wise, it allows them to stand. Where it finds them to be unwise, it exercises a veto power. It thus acts to protect management against the consequences of its own mistakes.

If there were assurance that the business judgment of commissioners would be superior to that of managers in more than half of the cases (weighted by their importance), we might conclude that duality of management would produce a net gain. But commissioners, in fact, are unlikely to be the better businessmen. And even if they were, there would be offsetting costs".

8 There have been exceptions to the rule that one only regulates where there are economies of scale and scope, but the outcome has been widely recognized as inefficient and inequitable. For example, motor carriers were regulated partially in response to pressure from already-regulated railroads. The result was to distort inter-modal competition and to regulate an industry, trucking, that did not share characteristics normally associated with market power. In the end, surface transportation was substantially deregulated in the 1970s and 1980s.
9 A self-regulatory organization ("SRO") such as NYSE considers the interests of other stakeholders as well. Traditional models of regulation, simplified to include "consumers" on the one hand and "producers" on the other hand, also result in rates and other relevant terms that balance consumer and producer interests. We discuss this in detail in our attached document on the Economic Perspective.
10 SEC Concept Release, op cit., p. 3.
11 SEC Concept Release, op cit., p. 3.
12 See Table 1 in our report, "The Economic Perspective on Regulation of Market Data Prices", for examples of vendors packaging real-time or delayed data.
13 Delayed data has been provided free by the exchanges to everyone for some time. This is important to keep in mind. It is not obvious that all or even most retail investors need "real-time data" in order to manage their portfolios, although they might have that preference (which they can exercise at a very low price, or, in many cases, at no cost at all).
14 This issue is fully explored and explained in our paper under the heading: "Regulation - The Economist's Perspective". In layman's terms, attempting to regulate on the margin forces a firm to submit to "death by a thousand cuts" as one service after another is priced at the margin and full costs are never recovered.
15 This competition continues to become more intense. Recently, the Pacific Exchange combined with an electronic communications network, Archipelago, to create an electronic stock exchange for U.S. equities. According to a Financial Times' front-page story:
The Archipelago exchange is expected to trade stocks listed on the New York Stock Exchange, NASDAQ and the American Stock Exchange.

It will compete with NASDAQ and the NYSE for new share listings, especially in technology start-ups. In the longer term, the partners expect the exchange to form the basis of a 24-hour global trading system. ...