AGS Specialist Partners
86 Trinity Place
New York, NY 10006

February 12, 2003

Mr. Jonathon G. Katz
Securities and Exchange Commission
450 Fifth Street, NW
Washington D.C. 20549

Dear Mr. Katz:

I am writing on behalf of my firm, AGS Specialist Partners, and I appreciate the opportunity given to comment on this proposed filing. Our firm, based on the American Stock Exchange, is the largest independent specialist firm in the industry, and I am writing to say that there are several issues raised in the filing that we feel warrant our comments.

Increased competition, implemented properly, can work to improve the liquidity in the current market, creating a better situation for all. However, when competition is applied in an inequitable fashion, it can be detrimental to the market at large. While there are many issues in the filing that concern us, the key issue is really one of transparency, and its effects upon the liquidity of the listed equity option markets. A lack of transparency would serve not only to weaken the market structure of this proposed exchange, but also to inherently weaken the markets in general, as other exchanges are forced to compete on what becomes an un-equal playing field. The structure set out for the BOX seems specifically designed from the ground up to facilitate internalization. The BOX has taken what is a powerful concept, that of open access and low barriers to entry (which we applaud), and added what is basically the crossing network of an internalization engine. This serves to undo all of the positive aspects (and then some) provided by their proposed liquidity-enhancing features.

Clearly, three seconds is an absurdly short time for price improvement. The BOX seems designed to disadvantage those who might try to improve prices, and thus prevent internalization, given the fact that a participant has such a short amount of time in which to decide to improve, and enact said improvement. With such a short time threshold, one can imagine that perhaps computer algorithms might become the de-facto method of bidding or offering during the PIP, but wouldn't this essentially mean that everyone disseminating markets is broadcasting an advertised price while secretly willing to pay more to any order that a firm might want to internalize? I would imagine this sort of anti-competitive practice would sound familiar, ie - "I'm showing a 2.00 bid, but I would really pay 2.09, provided my bid is not hit directly, but rather the order is entered into PIP, something to which not all orders have access." Additionally, the OFP seems to be given a second look, being allowed to tag along the price-discovery process as a lowest-common-denominator, but be guaranteed participation if they choose, leaving them scarce incentive to improve prices, and a tremendous incentive to try and act in their own interest rather than in the interest of their customer.

Why can internalized orders trade at prices that are unavailable to the market in general? In forcing the general public to bid and offer in larger increments than those at which the market-makers, and in particular, the orderflow providers, are able to trade, an inequity of opportunity is created. Why should an orderflow provider, charged with a fiduciary responsibility to their customer, be allowed to trade at prices that essentially "step in front of" a resting customer limit order? The customer with the resting order is basically guaranteed not to fill the order until market conditions make a purchase or sale undesirable to a professionally entity. To illustrate some of the problems, imagine this scenario:

A public customer is bidding 3.50 for 100 calls, which is the best bid reflected on any of the six options exchanges. An OFP receives an order to sell 50, and elects to internalize the order, entering the PIP process and offering price improvement to 3.51. This happens 3 times, with 3 successive orders of 50 contracts.. The customer bidding 3.5 has yet to purchase a single contract, and will be thusly unable to do so until the internalizing entity decides that 3.51 is not a price worth paying. Only then will the customer bidding be able to fill any of their order. The customer with the resting order was bidding the highest level anywhere, yet when 3 interested sellers came in, selling more than enough contracts to fill the order, the customer was not able to buy a single contract, prevented by an entity that is not representing any bid in the marketplace, and therefore not assuming any risk at all. If this were to keep happening, then the customer would have no choice but to replace the limit buy order with a market order to buy, thus paying the offer (at the very least 10 cents higher, but very possibly much more). The customer, who may very well have been willing to pay slightly more than 3.50 anyway, but is unable to due to lack of access to the same pricing increments, has now essentially paid at least 9 cents more than was necessary-- or fair. Essentially, the internalizing firm has profited by 9 cents (or perhaps even more!) at the expense of customer who was bidding at the risk of the market.

Perhaps even more jarring to the current structure and stability of the listed options market would be the ability of firms to pick and choose which orders they deem desirable, and internalize them, while electing to send undesirable or "risky" orders to other liquidity providers. This necessarily decreases the total available liquidity in the marketplace, as those who are providing the real liquidity to the marketplace only have access to those trades that others do not want. This kind of competitive disadvantage, built into the structure of the BOX, would serve to drive those willing to accept real risk out of the marketplace, unable to compete with those entities who control the orderflow, and can pick the trades in which they desire to participate, with next to no obligation or risk to the market.

We are already in an environment where the forces for internalization are willfully decreasing transparency. Over the last six months in particular, we have seen all of the positive effects of multiple-listing eroded by internalization and payment-for-orderflow. Previously, the customer's order would be routed to the exchange making the tighest, deepest, market and providing the best service (essentially, a race to the top). Now, in the current environment of internalization, there is a movement away from liquidity, to exchanges not willing to participate at competitive prices (a race to bottom). Orders which would have been directed to the strongest market centers are often not even quoted with those market centers, so as to prevent them from improving prices and thus impinging upon the firms' ability to trade against their customer at a price which is advantageous to the firm. The BOX, as described in this filing, while containing some powerful and interesting innovations, works too far to the end of furthering internalization. This will weaken, rather than strengthen, the competition and quality of the US listed equity options market.


Marc J. Liu
Options Specialist - AGS Specialist Partners
American Stock Exchange