February 7, 2003
Ms. Annette Nazareth
Re: Branding NYSE Liquidity QuoteSM: One Thin Line
As you know, the New York Stock Exchange is testing a market data and order execution product branded as "NYSE Liquidity QuoteSM". As a data product, Liquidity Quote joins our year-old NYSE OpenBookSM - our branded, publicly-disseminated limit order book - in providing institutional investors and their traders with information about supply and demand outside the BBO.
For each active stock, at a price point below and above the best bid and offer ("BBO"), Liquidity Quote will disseminate a firm "liquidity bid and offer" ("LBO") compiled from the limit orders on the NYSE book, from agents representing institutional orders on the Floor, and from the specialist for his or her own account. An institutional trader will be able to execute against the LBO through Institutional ExpressSM (IXPSM) or via a floor broker.
By filling in the price points within the 1/8th (12.5-cent) spread of the last century's BBO, decimalization reduced the BBO spread in our active stocks to a couple of pennies and made the "size" market opaque. Liquidity Quote recreates virtually the obscured 1/8th market, restoring size transparency and facilitating trading in sizes larger than typically found at today's BBO.
But Liquidity Quote does more than merely return the "size" market to the status quo ante. Our "live ammunition" training exercise with Liquidity Quote over the last week has elicited 1/8th markets with significantly greater size than the 1/8th markets of a few years ago. For example, the LBO spread in Texas Instruments on Wednesday was 11 cents, 33,500 shares X 50,500 shares. (The NYSE BBO spread in TXN averaged 1.5 cents, 3500 X 5300.)
Like OpenBook, Liquidity Quote responds to our constituents' desire for depth-of-market data. Like OpenBook, by enhancing the quality of our market data, Liquidity Quote increases the benefits to our constituents. At the same time, our initiative advances the public policy goals of enhancing market transparency, facilitating size executions, and fostering fair competition among orders, markets, data products and data providers.
The NYSE wants to compete by producing and disseminating differentiated, useful, quality market data. For us to reap the order flow rewards from our innovation, investment, risk-taking and quality assurance, our customers must be able to distinguish our data products and to strongly associate them with the NYSE. This requires distributors to respect the unique attributes of our product and brand by maintaining the integrity of our data compilation "package" and thereby assuring that our product is uniquely identified to the NYSE.
By definition, commingling products of two or more competitors in the same package, or mixing their products' ingredients within the same unit of consumption, submerges or obliterates each competitor's brand. It also has an anti-competitive leveling effect - it enhances the market position of the weaker brand at the expense of the market position of the stronger.
For vendors, interjecting other exchanges' data into the Liquidity Quote window is good business: it marginalizes the exchanges in the data business by relegating them to supplying raw material to vendors and by displacing the exchanges' brands. This protects vendors' existing franchises by excluding exchanges from competing in the market for finished data products. Moreover, for those vendors that also operate order-routing and execution services, displacement of exchange brands, aided by soft dollar arrangements, helps channel order flow to their ECN affiliates.1
OUR "WINDOW" REQUIREMENT
Our investment of human capital and financial resources in developing Liquidity Quote was not for the purposes of enhancing other exchanges' competitive position, ceding yet another market for finished data products to vendors, or promoting the business models of their ECN affiliates. So, as with OpenBook, we effect our differentiation and branding by requiring vendor displays to "draw a line" between other exchanges' data and Liquidity Quote.2 Our "window" requirement precludes a vendor from arraying rows or columns of data so that other exchanges' data "play off" of the NYSE LBO - from "injecting" other exchanges' data into the same "package" with the NYSE LBO. It also precludes a vendor from aggregating shares sought or offered on other exchanges with those in the NYSE LBO - from "mixing the ingredients".
Our window requirement does not impose rigid specifications. Rather it establishes a performance goal of independent display of our depth data. In keeping with the way we have administered the display of last sale and quote data for decades, we ask each vendor to provide us with screen shots and technical information so we can work together to achieve data independence in the context of changing technology and the rigidities of particular vendor systems.
This flexible approach recognizes that, to serve our buy-side and sell-side customers, we must minimize any loses in display utility while achieving our branding goal. But this is not a zero-sum, two-dimensional task. While at first blush, display independence seems at odds with utility, our discussions with vendors have shown that an imaginative use of our marks with screen format, lines and colors can strike an optimal balance that achieves both goals.
Thus, our window requirement leaves unfettered a broad spectrum of vendor displays. A compliant display can literally consist of a single row or column of our data among rows and columns of other exchanges' data - so long as the other exchanges' data do not "play off" of ours. Compliance does not displace the vendors' marks - it only keeps their marks from displacing ours. Moreover, the requirement does not apply at all to "end-users" that take the data-feed - broker-dealers, institutions and other non-vendors. And it in no way inhibits any competitor from disseminating its own depth data based on whatever value its data may have independently of ours - or from offering execution services derived from the NYSE LBO.3
We drew our window requirement as narrowly as possible, carefully targeting it to do no more than effect the differentiation and branding of our data product. Despite extensive conversations with the major vendors, none has suggested an even less intrusive way of achieving that goal.
WAL*MART CANNOT MIX BRANDED ANALGESICS IN SAM'S BOTTLES OR CAPSULES
Our window requirement complies with intellectual property ("IP") and fair competition laws, and comports with those laws' goals. Consider the case of a store stocking branded analgesic products.
Our laws preclude Wal*Mart from dropping a few capsules of generic analgesic into a bottle nearly full of "TylenolSM" and pasting on a "Sam's SM" label. Wal*Mart also may not open each Tylenol capsule, drop in a few grams of generic analgesic, erase "Tylenol" from the capsule surface and dump the altered capsules into a "Sam's" bottle. In the same way, our window requirement precludes vendor arrays that genericize the NYSE LBO, and prevents a vendor from aggregating the NYSE LBO shares with shares in quotes from other exchanges.
Note that, in the "array" example, Wal*Mart deprives Johnson & Johnson of the branding power of packaging. At best, if you can find your reading glasses, you may be able to make out "Tylenol" on the vast majority of the capsules. In the case of a "consolidated quote", Wal*Mart completely obliterates the "Tylenol" brand. Your reading glasses are no help at all.
Why do our IP and fair competition laws preclude Wal*Mart from burying or obliterating the "Tylenol" brand? Because mixing either the capsules or their ingredients enables the competitors of Johnson & Johnson to leverage off its innovation, investment and risk-taking in creating Tylenol. Mixing also puts Johnson & Johnson's product quality and reputation in the hands of its competitors. And it leaves only the "Sam's" mark in meaningful view - or in view at all.
In the same way, our window requirement prevents our execution competitors from leveraging off our innovation, investment, risk-taking and quality assurance in creating Liquidity Quote. Our window is our package. Customers can differentiate our data, consider our reputation and judge us on the basis of the utility and quality of our data product. It prevents deception by intermixing our firm, executable depth quotes with retail-size autoquotes. And it prevents vendors from "stickering" their logo over ours.
However, IP and competition laws do not permit Johnson & Johnson to interfere with Wal*Mart's display of its competitors' bottles on shelves over or under, or to the left or right of, the Tylenol display. Wal*Mart can also place Tylenol under the castor oil while shelving the generic analgesic in the checkout lanes. Indeed, Wal*Mart can feature other analgesics and not carry Tylenol at all. Likewise, other than by "drawing a line", our window requirement does not interfere with a vendor's allocation of its screen "real estate".
IP and competition laws also do not permit Johnson & Johnson to stop a consumer at home from mixing Tylenol with other analgesics, whether by mixing capsules in the same bottle or by mixing ingredients in the same capsule.4 Likewise, our window requirement does not preclude an institutional investor or its broker from creating "in-house" a montage that commingles data of our execution competitors with the NYSE LBO or from consolidating "in-house" the NYSE LBO shares with shares from other exchanges' BBOs.
But, insofar as an end-user acts like a vendor (i.e., provides data displays to its customers), its screens may not commingle data of other exchanges with our LBO. Insofar as a vendor or its affiliated ECN acts like an end-user (i.e., displays our LBO on its own employees' terminals), its screens may integrate our competitors' data with our LBO. If either a vendor or end-user becomes a "member" of the other's "class", we treat them the same.5
THE DISPLAY RULE: LESSONS FOR DEPTH DATA FROM A MISADVENTURE IN SOCIAL ENGINEERING
By downsizing the BBO spread to a couple of pennies, the Commission's decimalization initiative recalibrated its display rule to only mandate the consolidated display of retail-size quotes. This action aligned the rule with the protection of the class of investors that has the "first call" on the Commission - individual investors. Thus, without taking any further action, by requiring decimalization, the Commission gave itself a "blank sheet of paper" as to whether and how to regulate LBOs.
The Commission's Advisory Committee on Market Information (the "Seligman Committee") thought quite a bit about what the Commission should do with its blank sheet of paper in the context of its national market system mandate. While the Seligman Committee split over whether to require markets to disseminate depth data (e.g., whether to impose a firm depth-quote rule),6 the Committee never seriously considered extending the display rule to depth quotes. No Committee member argued that depth quotes provide an execution quality benchmark for individual investors - the key rationale for the Committee's lopsided vote to retain the display rule for "retail" quotes.7 Indeed, no member suggested that depth quotes provided any kind of direct utility for any but the most sophisticated of individual investors.8 In addition, the Committee believed that depth data should be relatively free from regulatory intervention. Like execution services, depth data should become a field for intermarket competition.9
Keeping in mind the Seligman Committee's premise that these distinctions between "retail" and size quotes should give the Commission pause before intervening as to depth data, please consider the following:
Finally we ask that you consider whether now is the time for you to consider intervention.
As was the case with Open Book, letting Liquidity Quote go forward cuts no irreversible channel. Moreover, the Commission will have as much jurisdiction in six weeks, six months or six years as it has right now. But, some day, the question may no longer be theoretical. Some day, the Commission will know whether any other market has followed our lead with executable, firm, non-fleeting, competitive liquidity bids and offers.
The matter is simply not ripe.
CAN EXCHANGES OFFER BRANDED DATA PRODUCTS?
We accept that regulatory intervention may be justified to assure continued dissemination of the NBBO to individual investors. But the competitive costs are staggering. Without evoking meaningful bids and offers from the periphery markets, the display rule's integration of the quotes of the Regional Exchanges and the Third Market with the pre-existing NYSE quotes genericized the data.
Today, as then, the display rule expropriates market data as a matter of branding, breaks the identity between us and our data, submerges our century-old data brand, masks our marks, erodes our quality brand through association with valueless data, creates a brand subsidy from us to our competitors, steals industry capacity, ejects the exchanges from the market for data products, and turbocharges the vendors' brand displacement strategies. The rule also extends and perpetuates the data consortia in enabling the Regional Exchanges and the Third Market to collect tribute for data that presents no value proposition. This tribute funds the rebates and payment for order flow that war against a brokers' duty of best execution.
In contrast, depth data is not about protecting individual investors. Recognizing that distinction, the Seligman Committee recommended that the Commission permit market forces to foster development and proliferation of depth-data products, and thereby achieve the NMS goals of greater pre-trade transparency and competition among orders, markets, data products and data providers. If the Commission wishes to heed the Committee's recommendation, it cannot divest the exchanges of the ordinary protections afforded by intellectual property and fair competition laws.
As the "Ellis Island" of markets, we urge the Commission to heed the Seligman Committee's recommendation. We want to enhance pre-trade transparency by delivering differentiated, firm, executable LBOs to the point of order entry - and to have the chance to receive size order flow as the return on our innovation, investment, risk taking and quality assurance. No Commission rule compels us.
We want to compete on our data outputs as well as on our execution services. We want to distinguish ourselves from our competitors through our depth products - not be homogenized with them. We want to compete with our competitors - not to consort with them, subsidize them, or otherwise give them exposure and credibility they cannot gain independently.
Otherwise, why innovate? Why invest? Why take risks?
* * *
I hope you found our recent meeting productive and find this letter to be so as well. We recognize that each issue can be explored in more depth. To that end, we look forward to meeting with you again on Monday.
Let me close with this observation. I drafted this letter working in a WordSM window. Several other branded windows - LotusSM, ILX SM and CNN SM - were running on three sides. As I named and saved my first draft, here's what "magically" appeared on my screen:
cc: Chairman Harvey L. Pitt