Ray,

Can you please explain how it is possible floor brokers contribute to price improvement. I have not seen it once. It may seem that price improvement is occuring if a floor broker is hiding his size at a different price than the shown bidd or offer, but I do not consider this price improvement.

by Ron on April 2, 2007 1:41 PM

Ray, I think the bottom line is that the original idea of the "Hybrid" was really great but it hasn't played out that way and we need to adjust it. I read a story from one of your links that said the specialists had given Mr. Thain an ultimatum. Basically if there isn't some changes in the way the current Hybrid works they are packing up. They find it too complicated and "Illogical" was the term they used. If the specialists are having problems can you imagine what the public is going through? I also heard that Mr. Thain is leaving for another position so maybe the specialists are not his main concern anylonger? Whatever the case those of us who rely on the NYSE to do our business need changes fast, like price improvement & matching, the cornerstone of the exchange. Thanks

by tony dey on April 2, 2007 2:19 PM

I think that's a great suggestion, but Ray pretty much sums up why it won't happen.

But then, there's been discussion on this blog in the past about "dark pools" and large liquidity elsewhere. How are they able to meet Reg. NMS?

I wish there was just one tape to watch. With all the ECNs, I don't know where to look for the best quotes...

I have noticed that lately I'm getting some improvements. Normally, they are for a penny or two, but at least it's something.

-DT

by DT on April 2, 2007 2:44 PM

Leave default electronic trading as it is and add an order type: "Route to NYSE floor". That wouldn't violate reg NMS mandate, would it?

If I wanted to execute a very large order, I'd rather pay a NYSE floor broker a little something to "work" the order. It might come up cheaper than the ECN's and be less work for a better average price. After all, floor people are the pro's. And maybe the "whales" would like the option too.

I do prefer the electronic thing, it works better for me, but there is no reason to throw floor people away. And adding that type of customer-selected "floor routing" would not remove anything away from the current system: only improve it.

The electronic system works a lot better for me in several ways and my bottom line has improved. I am increasing my listed volume. But I'm a small size trader (maybe that's why it works for me).

I'd like to see the other guys have the options they need to trade their own way and keep up their volume. Anything that decreases liquidity is bad.

by Angie on April 2, 2007 9:53 PM

Who is REG NMS allegedly trying to help? We know who it does help, but what problem was this originally designed to solve?

by John on April 3, 2007 7:35 PM

Ok, nevermind, I found the answer:

"...the NMS is designed to achieve the objectives of efficient, competitive, fair and orderly markets that are in the public interest and protect investors."

page 6
http://www.sec.gov/rules/final/34-51808.pdf

by John on April 3, 2007 8:03 PM

ok, so, i guess on those sets of stocks that fall into my criteria, it would be better to have .80 cent spreads, huge amounts of volatility, no matching, no price improvement, 0 specialist involvement (because they cant figure it out either) Well done. Good system. (Sarcasm intended)

Also, just heard the Thain/Euronext interview...He seemed to suggest more specialist involvement by implementing another "hybrid system" currently in the works. Any info on this would be greatly appreciated.

Thanks Again,

jt

by jt on April 4, 2007 3:21 PM

Ray, could you please give us some more info on what JT posted about Mr.Thain comments about a diff. or another "Hybrid" system with more specialist involvement? Thanks as always.

by tony dey on April 4, 2007 10:11 PM

JT, Tony -- I'm checking on this and will come back to you as soon as I get something.

by Ray Pellecchia on April 5, 2007 12:22 PM

I think this is what Thain's talking about. This is an in-the-words-of-Lou-Pastina-article that Wall Street Letter did back in January. These changes are being made now.

WSL: How will you go about speeding up hybrid and when will it be complete?

Pastina: One of the things that we are going to do is customize the point of entry into the total NYSE Group complex. Whether you are trying to reach the hybrid market or the all-electronic NYSE Arca platform, we are going to combine the two points of entry. The [current] options will be merged into a Common Customer Gateway, so you will have one connection into NYSE Group that will get you to either of the two marketplaces.

The second part is more dramatic. Everybody on the Street knows Super DOT, but they know it as a service. Super DOT is actually a system at the NYSE that sits in the order path before the NYSE's trading engine, which is known as Display Book. We are going to remove that from the order path and scale it horizontally with the trading engine to make a straight line from the Common Customer Gateway directly to Display Book. By doing that, we can probably reduce our latency [system response lag time] from 300 milliseconds down to 30 milliseconds. We would be looking to bring this new architecture online somewhere in the second half of 2007. Since Super DOT has been embedded in our system since 1978, it is a big piece of surgery to remove it and make sure that all of the things it does today in terms of reliability are still there tomorrow, but with improved latency.

WSL: Are you going to migrate hybrid technology to the NYSE Arca platform as Reg NMS is rolled out?

Pastina: We won't be putting hybrid on the NYSE Arca engine in 2007. There is some concern that when you do something like that you have to make so many changes to a system that is already successful that you wind up hurting that system. The NYSE Arca engine today is terrific in terms of speed. It averages somewhere in the neighborhood of three to four milliseconds. The thing I would hate to do is average it in with the NYSE so that the exchange could get to 10 or 15 milliseconds but Arca would be hurt in the process.

Sometime in the first quarter, if all the votes and required approvals go well, we will be faced with three matching engines: the Display Book, which runs on the NYSE, the NYSE Arca matching engine and the Euronext platform. We are going to have to decide which is the best matching engine and we will probably standardize on one, but I don't think that will be a 2007 exercise. What we will be doing in 2007 is evaluating how these run, both from a performance point of view and a cost point of view, and then in 2008 look at the next step, possibly, if it makes economic sense, to standardize.

by observer on April 5, 2007 12:27 PM

Dear Ray: Once again I face being the single optimist here. So far so good on the monumental changes at the NYSE. Yes, I'd like to see some additional price improvement but I think in time it will come as people see that hiding depth of their books only hurts them. I think what we should really do is have route order priority to those who disclose their true size. All exposed orders are executed on a first-come, first served basis and then the hidden orders. This way if you choose to hide your book, you pay a price.
As always.
Paulie Walnuts

by Paulie Walnuts on April 5, 2007 3:08 PM

I have previous about this exchange.Example in february it will go up...

by Ana Celina Toriani on January 7, 2008 2:14 PM