Eric,
I have long maintained that the whole fragmentation problem could have been avoided in the first place by making ECNs SOESable. That would at least put ECN limit orders on an equal footing with "live" MM quotes (i.e., those not in a 17 second dormant phase), and eliminated the old problem of SOES orders being rejected when no MM was at the inside market (the 90 second queue was a workaround for that problem). Had that been the case, SOES would never have become so useless, and it would never have been necessary to develop all the route picking skills and auto routing software we have today.
In spite of all the hindrances, ECNs have become the preferred execution path as long as their quotes are competitive with the MMs because they are the only place where you know the quotes are executable all the time. The problem with making them SOESable is the non-standard pricing structure which remains the domain of the individual ECNs. How can you do executions on a first ECN there first hit basis when one charges $1.00 and another charges $15.00 for doing the same thing? Most people still don't want to do a trade with an unknown cost.
I'm not sure if the ECNs can do without any fees, but there is probably a more transparent and uniform way for them to generate income. I agree with you that there is no need for MMs as we conceive of them, and that they long ago stopped performing a true MM function, but of course the market needs participation of a variety of entities that handle transactions for clients and institutions. (I don't think everyone can be a direct access trader.) Logic seems to lead to the conclusion that what we now know as MMs should operate exactly as the ECNs do with no dormant quotes, fully automatic executions, and perhaps even some compensation for performing the same function (assuming the ECNs continue to be compensated).
One of the things that concerns me is the change in order size from the old SOES tier limits, and the availability of the system to the MMs for their own trading. Whatever the motivation for the limits and passive MM participation in SOES, it did have the effect of increasing the chances of the small investor/trader getting the same prices as the big guys. I don't think we need a 5 minute rule, but some splitting and multiplexing of small orders with large orders might be needed to protect the smaller orders. I don't think anyone has studied this, and I'm just going on a hunch, but it I smell a gotcha coming down the road.
After reading some of the early central limit book proposals I came away with the distinct feeling that they were geared toward elimination of the ECN competition with the old guard MMs recovering the "lost" business. While the concept sounds great, somebody needs to stay on their toes to anticipate the consequences of the specific provisions. It's one thing to say we need uniformity and equality, and quite another to ensure that the people who have done the groundbreaking work of developing these alternative trading systems are treated fairly in the process, as well as the interests of all the buyers and sellers. BTW, I have no connection to any of them.
Dan |
