Date: May 26, 2003
Publication: InstitutionalInvestor.com

BUYSIDE TRADING ROUNDTABLE: Investors Say Many Trading Needs Not Being Met

A debate within the securities industry continues to rage over whether the structure of the trading markets is accommodating the needs of the institutional community. The advent of new order handling rules in 1997, the rise of electronic order routing, and alternative trading systems, as well as decimalization are all factors that have changed the face of trading. There has been a big push to boost transparency and liquidity, but this has also had some downside, such as the inability to get large trades done without compromising the interests of institutional investors, say . Wall Street profits margins continue to shrink, applying added pressure. And the New York Stock Exchange is conducting an investigation of some trading practices by specialists.

InstitutionalInvestor.com assembled an all-star panel of traders and market professionals to discuss the critical issues. In particular, institutional traders homed in on what they perceive as shortcomings in the listed markets. They debated a host of issues surrounding NYSE trading practices and structural concerns, including Institutional , , price discovery and best execution problems. As one panelist said, "We don't have the mechanism in place at the exchange whereby the best bid and the best offering is truly displayed for the investing public to make pricing decisions on their investments."

Richard Rosenblatt, president and ceo of Rosenblatt Securities, a floor brokerage, agency-only execution firm, joined the first half of the discussion to engage in a lively debate with the traders over NYSE trading operations, and points like the one above. "Whenever anyone bids or offers - they're really not advertising liquidity. They are representing something tactically in the hopes that they will attract liquidity to them. So as a tactical representation, I think that sort of runs counter to the transparency that, in the quote, I think you would like," said Rosenblatt, during one exchange.

The roundtable took place at with some participants patched in via conference call. The panel was moderated by Executive Editor Chris Gaudio and Associate Reporter Helen Shaw.

InstitutionalInvestor.com: What is the state of trading in the listed markets? Specifically what's your biggest concern, and what is the top priority that needs fixing?  Michael has improved considerably for electronic order routing. Things have really changed since the order handling rules went into effect in 1997 and all the competition from the ECNs. So Nasdaq has really improved for us in trading electronically, where New York really has not come up the curve as much. Decimalization has flushed it out even a little bit more. When we're posting a bid, we could get hit by a market order. If bad news hits the market, I buy the stock and down, that's fine. But in a stable market where a small market order comes in, someone on the floor or the specialist can price improve by a penny. And, of course, I'm not going to get called, 'Hey, do you want to price improve by two cents,' so that's a disadvantage for us. Also when we go to take liquidity, I go to take an offer, I'm pushing the button to take it, I made the move, but someone on the floor could participate with me or take it in front of me... 

We execute over 3,000 different trades a day, and I don't have the time to use a floor trader for every one. So I see it as a collision between the traditional auction market and electronic, and it just has too many disadvantages for electronic routing at this point. 

Richard Rosenblatt: Mike, you bring up some excellent points. I do penny on behalf of my clients. I've always felt that I don't have a right to spend any more of their money than is absolutely necessary to accomplish the goals that they've hired me for. The issue of the trading crowd and the inability for people off-site to access those markets and be ensured that the offer they're trying to take, the bid they're trying to hit is going to be theirs, is a difficulty that the exchange has wrestled with for a while. It's a very real problem. The way it was designed, any broker who has both sides of a trade must cross the stock, meaning that any other broker can break up that cross. It was never intended that the specialists would ever cross stock. When the [NYSE Super] DOT System and the Limit Order Book became so popular, the specialist was very often put in a position where he was crossing stock. And anyone in the crowd, myself included, would get a second guess and be able to break him up when he went to cross it. So someone who came into the system to simply take an offer ends up getting nothing, and my client gets to buy the entire offer. 

One way that they've decided to try to deal with that is Institutional LiquidityQuote will change that. I don't know if you're familiar with , but it is a pilot by the exchange to create a larger, more stable quote that won't be affected by the inside market changes due to decimalization. The advantage of accessing those markets through Institutional is that they don't have to be crossed. My clients' worst nightmare is that I expose what they want to do and they don't end up getting anything done. Institutional goes a long way to providing someone like yourself the ability to enter an order, to take a size offering, and if for any reason the offering has disappeared-been canceled or traded with prior to your order coming down to the floor-it's never exposed. Nobody sees it, including the specialist.

If, on the other hand, the offer still exists, then the Institutional XPress order would be exposed and be guaranteed no worse than the offering price. Any sellers in the crowd would be able to sell you the stock cheaper, again, providing me with the opportunity on the other side of the trade to penny someone. But the worst case scenario would be buying stock on the offer. There are a couple of flaws in the system, but it's a huge improvement over what currently exists. 

Buek: That doesn't solve my posting limit problem, though. If I'm posting that offer electronically, I can still get price improved and not sell the stock. 

Rosenblatt: Definitely. Keep in mind that the exchange is all about price discovery. It has never been structured as an automatic execution model. Actually, the only automatic execution on the exchange is NYSE Direct Orders, which the public really doesn't use very much. 

Buek: We love those. We use those a lot. 

Rosenblatt: Direct is useful for small orders where you want to access the market quickly, but are not available for price improvement [Editor's Note: Direct orders are only valid for order sizes of 1099 shares or less.] As they experiment with LiquidityQuote and automated quote changes, very often the automated quote is stale, the stock is trading possibly 10 cents above the offer, so anyone with the ability to take the offer in an automated system like Direct, gets a real bargain.

Kevin Connellan: I presume what Mike initially was referring to was Rule 76 whereby he would receive a "nothing done" when we attempted to take an offering electronically on the floor of the exchange. I still have issues, though, with . It certainly does help a little bit, but as I understand , the offering has to stay there for 15 seconds before it becomes XPress eligible. I'm sure John [Wheeler] would concur with me that 15 seconds in the life of a trader is an eternity. Certainly, the majority of our working orders are represented by brokers working in the crowd. But if I try to take 25,000 shares offered on the floor, via the DOT, my motivation is a speedy execution because I really want the stock. I certainly don't want the specialist, under the guise of price improvement, holding it up. And so consequently, even doesn't fulfill my needs. I think we've got to get the exchange to arrive at "full electronic transparency" where the first order in gets the stock, period.

John Wheeler: There's still some struggle at between venues that are posting markets that are not Auto Ex and those that are like the AMEX. They continue to quote these markets 5,000 shares up with a penny spread, and they're impossible to get to. So still some lingering issues there, but for the most part, the connectivity and the Auto Ex is there, so our focus has shifted pretty much entirely to the New York Stock Exchange where we see a multitude of problems. I'd agree with basically everything I've heard so far. I would probably dampen the enthusiasm that I've heard about and LiquidityQuote, because I agree with Kevin, 15 seconds is a lifetime. From what I understand from being at the exchange a few weeks ago, there is a rule change on the books to take that timeframe down to zero. But nonetheless, it still has to be for 15,000 shares, and it's only for the quote. It's only for the best bid and the best offer. It says nothing for other orders that may reside on the specialist's book, up or down just a few cents, for some really significant size that may be sitting there. 

Number two, on it's very ambiguous and arbitrary. It's kind of an amalgamation of specialist interests, floor interests and orders on the book. And it's entirely up to the specialist to post that price and that size. While firm for those who want to hit [that price], it's still subject to price improvement from the floor. They have been running a pilot of 400 stocks with posting, and [the question is] whether this is just a kind of spring training that will get better once the full season starts, or this is how it's really going to be. But from what I saw, standing at the Wal-Mart post after they opened the stock, LiquidityQuote showed 50,000 shares up/down 30 cents in Wal-Mart, but the floor brokers were walking in the crowd and buying 25,000 right in the middle of the spread, or selling 25,000 shares in the middle of a spread. So, as an institutional product, if you can buy 25,000 in line, or 50,000 up 30 cents, it's exactly like buying 25,000 in line and 25,000 up 60 cents. Unless they really tighten it up, I don't see that it helps a whole lot. Also, all of these initiatives from the exchange, in my somewhat cynical view, are really designed to protect the intermediaries. They protect the Dick of the world. I'm sure Dick's a nice guy, but the structure of the exchange allows a Dick Rosenblatt to stand in a crowd, as he just said, and penny everybody that tries to interact with the exchange. I see that as a huge problem. That's how Dick makes his living and that's well and good, but I don't view that as an efficient marketplace, because there is no incentive to display best price. The best bid or the best offering ought to be displayed for the investing public. 

[NYSE Chairman] Dick talks about the school bus drivers and the house painters of the world. Well, those people deserve to see the highest bid and the lowest offering so that they can make conscientious investment decisions with their 401(k) and their retirement plan. We don't have the mechanism in place at the exchange whereby the best bid and the best offering is truly displayed for the investing public to make pricing decisions on their investments. The data from the exchange tells you that 40% of the orders are price improved. They're proud of that. I think it's disgraceful. And if there's a better bid in the crowd or there's a better bid by the specialist, it should be displayed for everyone to see and we're not seeing that. But the bottom line, as an institutional participant at the exchange, the exchange has us in a corner. It forces us to choose between anonymity standing. You can have standing and be first in line and do the if you hire a Dick Rosenblatt to stand there and represent you. Or, you can have anonymity, but you go to the end of the queue. There's certain orders and certain stocks where neither one of those are very advantageous outcomes for my shareholders. I'll end my spiel with that. 

Rosenblatt: You really packed an awful lot in there. Keep in mind, when I trade over-the-counter I use every ECN that's out there, just about. I also trade on the floor, obviously. You're my customers. Basically, what I do with an order is what my clients would want me to do. None of my clients want me to expose the potential liquidity that their orders represent. Whenever I bid - and I think whenever anyone bids or - they're really not advertising liquidity. They are representing something tactically in the hopes that they will attract liquidity to them. So as a tactical representation, I think that sort of runs counter to the transparency that, in the quote, I think you would like. And it's that the customers, it's that the public, the institutional investors, don't want their bids and offers shown. 

Wheeler: Dick, I'm going to stop you there and just argue with you on one point. I would love to display liquidity on the floor of the exchange. There are a lot of stocks where I want to be able to show a 100,000 share bid and 100,000 share offering. The problem is that the structure doesn't allow me to do that without my interest being compromised. and LiquidityQuote [should be] turned around so that there is protection for the limit orders that are displayed. We always talk about how there's nothing on the book. New York publishes open book - gee, there's nothing out there. We're looking at it backwards. There's no incentive for anybody to display anything right now because all of those orders are subject to the auction market. They're all subject to price improvement. 

Rosenblatt: Right.

Wheeler: That will never change until we flip-flop the algorithm and say, 'When someone's bidding 20 bucks and somebody comes in to sell them at 20 trade happens.'

Rosenblatt: That's what an ECN does. And if the investing public, my customers, decide that that's where they want their marketplace to be, that's where it will be, and the New York Stock Exchange will cease to exist. 

Wheeler: We fundamentally haven't had a choice yet, Dick. The rules and the lobbying efforts of the exchange are such that there is no viable alternative. I mean, I love Mike Cormack to death, but until is fixed, I don't know how much volume we're going to be able to run through the P-Coast. But that's a model that I love, and that I think a lot of institutional participants would love to participate in it as well. To just say that 'This is how it is and this is how it's always going to be, and we've just got to live with it,' I think is wrong. It's wrong for our investors. We should be trying to make the markets better, not discovering ways to cope with the status quo. 

Mike Cormack: I think one of the points that Dick brought up and one of his quotes is that 'New York is all about price discovery.' I think the price discovery process works when you're on the floor in some respects. You have to be a member of the club, or have somebody who representing you down there. I think Michael was saying to a certain extent that's just inefficient and takes too much time. I would agree. In a market where somebody has to execute tens of thousands of trades a day or more, which we frequently see with clients, it's impossible to execute and feel comfortable that you're getting best execution if you can't send your order down there electronically and feel like it will be handled in a fair manner and not disadvantaged vis-à-vis somebody who's actually standing in a crowd down on the floor. 

And I think the second point which both Kevin and John have reiterated, that from our perspective as an electronic execution platform, we think we owe Michael everything for posting that bid for 3,000 shares on our system. He's first in line, bar none. Nobody gets any special advantages ahead of him. If anyone wants to step in front of his order, whether it's by a penny or more, that's at their risk. And Michael has the ability to cancel his order instantaneously off our system and remove the free option value of that trade... 

We have clients who have quantitative models that are really only valid for, say, seven seconds. New York takes 15 seconds to turn around a cancellation. Those models can't be implemented in listed trading without systems like ours. And what does that mean? Well, as people use Archipelago or other electronic platforms, they could implement these strategies, compress spread and provide more liquidity for the buyside or for retail or anybody else who wants to transact against those orders. 

Wheeler: Price discovery on the floor of the New York Stock Exchange is not just buying the stock a penny better than the other guy. There is, in essence, no price discovery on the floor of the New York Stock Exchange, because there is no reward for being the aggressor. There's no reward for setting price. There's only a reward for standing in the crowd and waiting for someone else to make a trading decision and simply stepping ahead of them and breaking them up. That's not the right price for a given volume of stock. It's a penny better than somebody else is willing to pay. And that's not right. It's not right for two parties to agree on a price and trade stock, only to have intermediary break that trade up. 

Connellan: For a penny. 

Rosenblatt: First of all, Mike Cormack, I think you've got a great product. As a client, I'm also a fan. But the reason I said that the New Your Stock Exchange is all about price discovery is it's the only reason that anyone uses the New York Stock Exchange. The model which you're complaining about - and you have every right to complain about it - is a model that is designed to have a negotiation, whereas you want an automatic execution. That's contrary to what the exchange is designed to do. It does that miserably because it doesn't intend to do it. It intends to have every trade have the opportunity of negotiation, so if anyone wants to pay more - and I do hear you about the penny - they have the opportunity to do that, so the resultant price represents the most anybody is willing to pay and the least anybody is willing to sell stock for, at that moment in time. I do agree with you. If all that was happening on the floor in terms of price discovery, those people standing around bids and offers, the price discovery would be very weak and I think the would grab the majority of the volume very quickly.  But that isn't what happens. Very often - actually most of the time - you're not pennying. Most of the time you are making trading decisions to take an offer or maybe just get out of the way if you think the stock is going to go lower. That is the heart of the New York Stock Exchange. Whether we like it or not, it is designed so that electronic order flow negotiates with traditional order flow. As long as clients validate the system by sending most of their order flow there, it's going to be preserved because people like are expected to take advantage of that price discovery mechanism for my clients. If my clients told me, 'Dick, just go take an offer. Just go hit a bid. All we want to do is ensure that we get a trade,' then the whole dynamic on the floor would change.

The other thing is you don't have two parties agreeing to a trade. When someone sends a limit order down to the NYSE, there is an assumption by most of the people that send that order that if in fact somebody's willing to pay higher than the bid, then they will get the benefit of the better price. It's an assumption that I understand you may not like, but it's still there. And to be fair to all the players, I think an awful lot of people like that opportunity to price improve. Of course, they would love it to be much more than a and I think very often it is. But, that seems to be why the business goes there. I mean it costs me a fortune to executive my clients' trades on the New York Stock Exchange. We have five seats. That's a lot of money. On the other hand, I get paid to make a difference. When I'm trading listed stocks, I simply can do a better job for my clients on the NYSE - often through one of my automated systems. In fact, we were the first firm to put a DOT System on a non-member's trading desk. So I'm not only a fan of customer DOT, but we created it. 

I'm a little frustrated because I understand your complaints. Except it is your colleagues, taken as a whole, that support and demand the system to be the way it is. 

Connellan: I'm not so sure it's entirely accurate that they do. I would go back to what John was saying earlier, it's just that there's no viable alternative out there at the moment. The Archipelagos and the are certainly making inroads, but there's still a long way to go there to get the market share away from New York. As I talk to more and more of my peers, it's clear to me that they believe that the model is outdated-the negotiated model-and we need to have a better electronic model down on the floor of the NYSE. It's also interesting to note that the themselves, even though they may endorse the negotiated NYSE, they still have the Millennium that they enter orders into before it goes down to the exchange and they do some electronic crossing there, for efficiency. 

Peter Jenkins: Dick's a floor broker and he's dealing with tools that have been given to him. I think the crux of the problem here is that the electronic system, which is Super DOT, is an order delivery system. The NYSE has not come up with a solution to electronically interact with the auction. I think Archipelago is trying to create a solution, which is all electronic. I'm not too sure that we can successfully, or the NYSE can successfully, mesh the two and maintain the auction. That's the bottom line. The NYSE is going through a very frustrating period, trying to preserve their auction. Yet, unfortunately, it's all about slowing the order down. The demands from the buyside are to make things more efficient. I've heard the term 'freezing the book.' That's like saying, 'Hold on everybody. Let's slow everything down here so everybody can make a decision before we actually trade something.' We're supposed to be moving towards efficiencies and speed in the marketplace. 

Wheeler: To me, freezing the book is an admission that a human being is not capable of handling the order flow as fast as it is reaching the floor. You never heard of freezing the book 10 years ago. 

Jenkins: That's right. John, it's their way of dealing with the heavier volumes. 

Wheeler: Sure.

Jenkins: Instead of putting in technologies to speed it up, it slows it down, and I think that's what we disagree on. We're not looking for the process to slow down...There's a tremendous amount of pent-up liquidity that could be in that marketplace but never makes it, because of the reasons John mentioned...The NYSE could probably be doing at this point three times as much volume, if it had the right technologies down there. 

Rosenblatt: Peter, I really hear you loud and clear, but I just want to say that you're defining efficiency as speed. The exchange, although it seems to be trying to make things faster, defines efficiency as price discovery. So that the goals here don't seem to be parallel... 

Jenkins: I'm not too sure their definition of price discovery is the same as their largest constituency, the institutional . The institutional description of price discovery is not saving a penny on 1,000 shares. It's moving massive amounts of stock in and out of the marketplace. 

Wheeler: When trading a listed stock, I always keep open book up on my screen. I had two situations where there was sizeable stock available on the book, up two, three, four pennies, and I sent DOT orders down to the floor to, in essence, sweep through those offerings and reach the highest price and the most liquidity and trade the biggest block of stock. In the two situations, the book was frozen and a larger piece of stock traded a penny higher than my bid. Now, to Dick I would say: How can that be price discovery? How can that seeking out the best price for all participants? I view it as protection for intermediaries. I don't know how else you can look at it any differently. Anybody who had a buy order on the floor in the crowd at that time saw those offerings, had an opportunity to buy stock at those prices - and because I try to deal anonymously with the floor through DOT, the book is frozen. Everyone in the crowd gets a chance to trade a penny above my bid and I'm left holding the bag. That's not an efficient marketplace for my investors. 

Rosenblatt: John, I hear you. I can't comment on a specific situation, since I don't really know what happened...My clients use DOT extensively. As I said, I have a DOT system that I give my clients. When something happens that they don't think is right, one of my brokers goes in and finds out what happened. In terms of freezing the book, the only time the book should be frozen when it's being used properly is when a trade has occurred and the exchange has automatic cancellation. So, if in fact... 

Wheeler: Dick, let me stop you there one second. You hit the nail on the head. The trade has occurred. 

Rosenblatt: Yes.

Wheeler: So the trade happens first. Then, let's freeze the book and rebalance. 

Rosenblatt: It's not rebalancing....When the book is frozen, this order that you sent in that you cannot participate on that trade, no one can see that, including the specialist. He may get an indication that he has messages pending while the book is frozen, but he can't see what they are. The reason that the book is frozen after an execution occurs is because the offers that have been taken are on the specialist's book. After the transaction if you send in a cancellation, the order will simply disappear and, in effect, you'll be walking away from a consummated trade, so the specialist freezes the book to give him the ability to report those trades to the sellers prior to their ability to cancel their offers after the trade has occurred. That's the only proper way that the book should be frozen. Anything else, you have a legitimate complaint. Send a broker - I'm sure you're associated with some brokers on the floor - send somebody in to check it out. 

Wheeler: But see, that's what I'm trying to avoid, Dick. There a handful of my orders that I don't want to be sending a broker into the crowd for. 

Rosenblatt: No, no. I'm talking about your DOT order. If you think your DOT order was mishandled, send your broker into the crowd to check it out. I'm sure you pay enough commissions on the floor that if you asked your broker to go check a DOT order that he didn't make a commission on, he's not going to complain about that. He'd probably be happy that you're depending on him. 

Wheeler: But here again, you're trying to rectify the symptoms. I'm trying to fix the system. 

Rosenblatt: What I'm saying is it's not intended to be an automated system. It's not designed that way. 

[Editor's Note: At this point, Dick Rosenblatt had to leave the discussion.] 

Connellan: It just seems too simple a premise for the NYSE to grasp...to just 'show the best bid, show the best offer,' and the size, and if somebody comes in to take it to allow the real buyer and seller to consummate the trade without an intermediary sitting down there trying to do it a penny better, even though that person may well be representing a legitimate customer. But he had his opportunity. To me, by standing in the crowd and not taking the offer, he's risking the fact that somebody else can come in to take it. It just seems a very, very simple premise, and I just don't know why doesn't get that. 

Wheeler: I'll add to what Kevin was saying by saying it would not take a whole lot of work on the New York Stock Exchange's part to tweak these initiatives to better accommodate the institutional community. And a couple of those would be any offering of 15,000 shares that's been there for 15 seconds  - or five seconds for that matter - should be fair game, because brokers in the crowd have had the opportunity to trade against that, to trade with that bid or offering. Once it's been there and it's on the book and aged, whatever the right aging period is, it should be available for anonymous DOT interaction by the institutional community. 

Bueck: We continue to let New York go along because you are advantaged. If you have a large order, you are advantaged to use a floor trader, to penny and get in front of electronic orders. When we have the situation when there's large orders in the order book, I'm forced - and I hate to do it - I'm forced to lose my anonymity, go to a broker and tell him, 'Hey, it looks like there's something there. Go ahead and get advantage by going down to the floor.' Where I'd much rather stay electronic. It's cheaper and I keep my anonymity. So we keep the game going. 

Wheeler: It's like an addictive drug. Everybody wants to do the and nobody wants to be . As long as there's a facility and [we're] allowed to do the instead of being , it's going to continue. But I think the institutional community has come to grips with the fact that, as a community, we are being compromised. a handful of trades in any given day where you are rewarded by standing down there and doing the . I think everyone realizes that the intermediaries are profiting far more than any of our good trades by the system that's in place on the floor of the NYSE. 

Bueck: And [with] the electronic orders, it's just not minimal any more. In the first quarter, I think 37% of the volume was electronic. 

Wheeler: Right. There has to be a facility for the electronic orders to have the same standing as a [specialist] in the crowd...There's something wrong with a system where a human being can access that stock on behalf of a client, but the electronic order cannot. It's the same people behind those orders. It's still the investing public that's behind those orders. The fact that they flash orders based on whether it's presented by an NYSE member or an NYSE proprietary product in DOT, is fundamentally wrong. 

Connellan: It's ironic, too, because some institutional investors have DOT machines that are sponsored by brokers or members of the exchange. 

InstitutionalInvestor.com: Is there a greater balancing act when trying to deal with the cost of trades? What are your thoughts on the current landscape?

Connellan: The landscape on Wall Street keeps changing every month. With all the layoffs, we've seen considerable shrinkage of talent. And not just on the upstairs coverage, but also even on the floor where many of the major firms have laid off senior floor brokers replacing them with junior talent. So execution is harder to come by and even though we are not necessarily dependent on capital commitment we see less of that also, as large sell-side firms move their resources away from their agency trading desks to their proprietary units. Price discovery, NYSE initiatives notwithstanding, remains elusive. 

Cormack: Virtually all of our direct clients are broker/dealers as an exchange, but we've seen tremendous price compression over the past 18 months with being pretty aggressive the first quarter of last year, and Nasdaq and Archipelago matching their prices. So there isn't much there. Now on the broker/dealer side it's very much a scale business. We've also seen, from a more anecdotal basis, a fair amount of price competition, particular in the hedge fund space for on the broker/dealer side looking to grow that business. Those guys seem to be typically more price sensitive than many mutual funds. Maybe they don't have as many soft dollar obligations. I'm not sure.  

We're also seeing, with decimalization, particularly in OTC securities, a tremendous amount of liquidity at really tight prices. We see typically more liquidity displayed a penny up in or OTC securities. We also see a lot of reserve size there, both at a penny and then up two or three cents. There's real liquidity - maybe 10, 20, 30, 50,000 shares, paying at two, three cents up, in a name like Cisco or Intel, which you don't see that as easily in New York with said names. Typically, they're 100 shares up. Sometimes two or three or four cent spread. OTC just seems a lot more efficient, and I would guess, even from an institutional perspective, cheaper, both explicitly and implicitly to trade in. 

Jenkins: Actually, on the explicit side there's something now that I find pretty interesting. The rate has been  generally fixed on a listed order, for a research type order,  somewhere around five cents. With this onslaught of direct access brokerage, I've been getting a lot of calls from the traditional full service research brokers. They're now starting these services, direct access services, under the guise of 'We want to offer our clients all of the products available out there,' yet they can't answer the very simple question: 'If I'm working 99% of my business as agent, why would I use your direct access at two cents for my agency orders, and just use your upstairs broker when I want to use capital?' I think they're going to get themselves caught here, because Lehman [Brothers], [Merrill Lynch], I think Goldman [Sachs], and Bear Stearns are now offering this direct access. So I find it interesting that through trying to capture market share they're essentially going to cut their listed commission rate now in half.

Bueck: We've used a lot of direct access brokers, even from the major firms you wouldn't expect to have it. And aside from the lower commission, you're much closer to the point of the trade, so you're getting better information quicker that you can respond to. So not only is it a third as expensive, commission-wise, but it's better quality execution.

Jenkins: I know that. But what I'm saying is they're offering you this at a reduced rate, which I find kind of humorous, because they're saying, 'Here you can get this better execution,' which we technically should have had all along. Why didn't we have access to these floor brokers all along? It's not like a new technology came out. We're still using the phone or fixing the orders through the brokers on the floor, yet they're charging us less than half to get their service. 

Connellan: And Peter, on top of that, they're cannibalizing the business even further by offering

Jenkins: There is a tremendous compression now in the explicit rate. 

Wheeler: I think what the bulge bracket firms have failed to realize is why so much business has gone the route of direct access. I don't know how the rest of you guys feel, but on our side it isn't to save two cents a share. The reason a lot of volume goes to direct access brokers is because you don't get all of the phone calls that you get with an upstairs trader. You don't get all the protection being offered other customers. The first call isn't to someone on the same side as you who traded them yesterday as a 'courtesy.' You don't get the trade building exercise going on...A Goldman Sachs or a Merrill Lynch or a Morgan Stanley doesn't want to make the other side of that trade look bad. I think that's a big reason that a lot of businesses move to the direct access route, away from the upstairs desks, is to get away from all of the shopping of order flow that happens on the sell side desk. 

Jenkins: The unfortunate thing is I'm seeing this trend of what I consider some pretty talented upstairs traders taking that same talent and just moving to the floor and making the sales trading call direct from there. So, I'm not too sure

Wheeler: But I think they're finally rationalizing their business. The brokers have rationalized their business whereby a dollar of commission isn't a dollar of commission anymore. It depends on what venue you use to pay them that dollar. They are, like never before, keeping track of every single contact with our portfolio managers, every single , every single company visit, every single conference, the participants, where they stayed, etc. The Street has finally realized that we don't need 150 analysts in Coca Cola, and they have made it a two-way relationship. We have always rationalized the services that we get from brokers. They're finally looking at it from their side and saying, 'Gee, what really paying us for? And if it's not for the 150th analyst in Coke, let's not have this analyst around any more.' 

InstitutionalInvestor.com: What are the long-term effects of this? How do you see a new landscape developing? And is it good or is it bad? 

Jenkins: It's hard to say right now because there's one important ingredient in the marketplace that's missing, and that's investment banking fees. Investment banking fees are down, what, 90%? 95%? At some firms they don't even exist. When that comes back and the volumes expand again, it will be interesting to see. I don't have a guess right now but certainly the execution side, they're battling over flow at this point. I don't know what happens. I don't know what the final outcome is...I have heard some unsettling news, just talking to people - you get some good information from people that have been laid off. I've talked to a lot of position traders and sales traders that have been laid off and one of the themes is: They didn't really like what they saw going on at their firms because there's a lot of build-up on the proprietary side. So I think not only are you seeing this in the specialist units on the floor, but I think until these investment banking fees come back, you're going to see a lot more proprietary trading build-up at the brokerage firms. 

Wheeler: I'd agree with you, Pete. I will make a prediction about what it looks like when we come out this and say, it will never look like it has in the past. Because I think when we do come out of this, commission rates are never going to ratchet back up. The only way to capture more percentage of principal is to keep splitting share prices and keep them all at $3 or $4 or $5 a share, since the two or three cents that we are paying is a minimal amount of money to the brokers. But when we do come out of this and the investment banking business does pick up again, there's going to be so much scrutiny on the allocation side of the IPOs like we've never had before. I just don't think we're ever going to get back to the excesses that we had in the 90's. The brokers historically over-fired the lows and over-hired the highs. But I think it's going to be different this time coming out. On top of that, we have we have the Archipelagos of the world and the other executing venues where I really see that the explicit cost of executing a trade approaching zero. 

Jenkins: I just hope that this bear market isn't so stretched out where the and the various that are spending money to build a model don't run out of money. Because I think people are still fighting for market share. But as Mike said before, the fees for doing this are being driven right into the ground. I think the only way that we can have substantial change going forward is if we do have these competitors out there. And I just hope we don't lose that. But if you're that concerned about it, I'm sure would sell you a 10% stake. 

(Laughter)

Cormack: Although the margins are tight, we are profitable. You can make money in this environment. I'm not saying you could make a lot of money. But what you have seen in our space, and frankly you've seen it in the industry in general, is the consolidation and the economies of scale, both with the larger broker/dealer side, the money management side, and certainly in the ECN space. I would expect that to continue to a point. I don't know what the right number of liquidity pools is. We haven't talked about fragmentation yet, but there's always a lot of concern within the industry about that. But the flip side is that Archipelago, , New York, are out there fighting trying to kill one another every day, and I think in many respects that is good for you guys, and it's good for investors in general. Because it's brought on a tremendous amount of competition. If just one of us left as a de facto liquidity pool, I think the only way you can control that kind of monopolistic behavior is through more regulation, and that's not always so efficient for getting you guys the features that you want and the pricing that you want. 

InstitutionalInvestor.com: What about fragmentation? 

Bueck: The fragmentation doesn't bother us at all. The competition has brought all the changes, I think without the competition we'd be in pre-97 trading again. So now with electronically being able to connect to the fragmented pools, the huge gains you've got with the competition, the little bit of extra effort to make sure you check the pools is not an issue at all. 

Wheeler: The fragmentation is not the I think the issue is connectivity. As long as you have efficient lines and sufficient market structures, you could have liquidity pools all over the globe. I'd rather try to interact with a market maker in New York than try to interact with the AMEX on an OTC stock. Because of the market structure and the connectivity issues there, and the lack of Auto Ex. So, it's not so much that of liquidity out there, as long as they're connected efficiently and the structures allow for Auto Ex, it's a non-event. 

Connellan: I'm not as concerned about the fragmentation, given the fact that we have the connectivity to all of these various , etc. The only thing is that down the road, of course, once you've been doing it for a while, you probably will want to see a little bit more of a cohesive effort. It's okay when you're taking stocks and you're taking them from the various ECNs, but then the time eventually is going to come when you're sitting there with an order where you've taken everything you can find at your price and now you're bidding. Then you've got to make a decision about where you're going to be represented. And obviously, you can split the order and do it on different [markets], but it still requires a little bit of effort, and I'm sure somewhere along the line somebody is going to try and find a way to automate that process. If you are bidding on multiple platforms and keep receiving reports from one particular ECN, the algorithm should automatically take orders away from the other and keep feeding the ECN where you are having success. 

Cormack: I have a question for you guys. We recently moved out of the as we the OTC symbols into the exchange platform. We currently route all the other liquidity pools. So we route to SuperMontage, we route to Instinet, Island and Brut and B Trading, even though that's . So, on all marketable orders we're able to go outbound, and obviously if you post in our system you're displayed in the national market system.  But a client said that they didn't want to post on Archipelago any more. They preferred a post on Instinet will go get their order. But if they post on Archipelago, wouldn't necessarily go access our order, on our system. Because doesn't do the routing like we do. 

Connellan: Yes. There's been occasions of locked markets. We encounter that all the time. With the locked market, that only happens with Instinet. At times what we have to do is we have to resend the order, basically. 

Buek: You have your choice. When I go to take liquidity out of the system, I'm going via the smart router. And, of course, going to access its book first, but then it's going to access ARCA, Island - the whole gambit. You do have the ability to lock a market, if you want. From my viewpoint, I have no reason to do that. Potentially there are some traders out there that are trying to get the access fee. So they'll post, lock a market, hope someone hits them, and then they'll get paid the access fee. But I'm on to trade, so if I want to take an offer, I'm not just sitting in the book and locking the market. I'm taking everything. You have the ability to do both. 

Cormack: My response to that was, obviously tongue-in cheek, 'Well then, we'll just stop routing Instinet and you'll want to post on our system again.' So I mean it just didn't make sense to me why we would be penalized because we're trying to and route, versus some of the other firms, our competitors who don't route like we do. 

Buek: Mike do you allow clients to lock a market if they want to? If has a 20 offer, can I post a 20 bid in ARCA?

Cormack: Yes. If you use a post no preference order, you can. But the vast majority of our clients don't do that. They just use our normal routing algorithms, which always route out. Now in the one instance, too, where you mentioned that people are going for the liquidity rebates, we were actually seeing a lot of that. What people would do is they would post an order on Island to get a liquidity rebate and lock across us, knowing that we would activate and then go and then go get that order, which obviously has cost for us. But it has cost for the client too, because there's a pricing difference for a broker/dealer when we go outbound and hit or take Island. So they were in effect stripping those orders off our book. So, what we've been doing since that's been discovered is we now stand our ground. So if we're there and somebody proactively locks us, we do not go get that order. So it is possible for clients to proactively lock across using that post no preference order, but it doesn't happen a lot. 

Buek: As your client, I think I would like you to take Island. And the reasons are lame, but if they're posting at the same price that I'm posting, I would like you to go outbound and take them and hopefully work out the issues. 

Cormack: The problem with that is that they basically strip our book, and then the client on our end is charged more money for accessing that liquidity. It's we could just do a different order type that would activate. If there was demand for it. We'd price it differently. But it's doable. 

Wheeler: We could spend three hours on access fees and liquidity rebates. My guess is there's some of that behind these people that want to post on only. Another facet of that might be the sub-pennies that Island allows you to post and some of the other like Archipelago don't. 

Cormack: Our perspective on the sub-pennies, we actually did offer them in the ECN, and we don't offer them in the exchange any more. The vast majority of our clients don't like sub-pennies. I'm guessing that most of the doesn't like them either. The other issue, which was kind of surprising, is that the guys who originally were out there breaking the penny and making quantitative markets, the profitability of those has shrunk so much now that everyone's tenth of a one another on these other systems. Many of those clients told us to stick with a penny. From a business standpoint, that's really not the hill we're going to die on as an exchange. If we had a huge amount of customer demand, we would talk to the SEC about going sub-penny, but I imagine that's going to be a pretty high beta issue for the industry, and it's not something that we see enough demand for. So we're pennies, although we will go outbound if we see Island or Instinet at sub-penny increments better than our own book.

[Editor's note: , which was not represented at the roundtable, made the following comments when subsequently contacted by InstitutionalInvestor.com. Michael Plunkett, senior ., managing director, hedge fund segment at , said the ECN gives clients the choice to route however they want - all over the Street, or in native liquidity pool. It's a function of clients' behavior and they will decide when it's best to route or to stay on , he said. "Locked markets happen throughout the market." Locked markets depend on trading strategies and clients rather than a particular ECN or broker, Plunkett added, noting the only they can't pro-actively route out from is SuperMontage because doesn't route out to other liquidity pools. Locked markets are unique to a client strategy.]

InstitutionalInvestor.com: How has changed the market? Is it working the way everybody thought it would?

Connellan: I think that the short answer is it wasn't the roaring success that [the NASD] had hoped it would be because the others are still around. I'm sure that they'd even have to admit they have not garnered the market share that they thought they would, given

Wheeler: It's just another ECN, in my view. Had it rolled out two years earlier, maybe it would have made a bigger splash, but it's basically a non-event. I guess the only reason I can really come up with is when you take a bunch of very competitive broker dealers and you tell them they have to use one venue, they feel like they are really playing into their competitors' hands by hitting and taking their competitors' bids and offerings. A lot of the broker/dealers in ECNs to trade with natural counter-side liquidity, rather than achieving the objective of their competitors; i.e., executing other people's orders for them. 

Buek: It's turned out to be a non-event to us. All the market participants have gotten very used to getting their down before SuperMontage went live. So I was initially afraid that possibly there'd be some trading around my [other ECN] bid or offer, but it turns out that since all the connectivities have been established, it really was a non-event.

Jenkins: I actually just have a question. Outside of ECN access to SuperMontage, have any of the traditional houses put access on desks to SuperMontage?

Connellan: Obviously Goldman has been vocal about doing that with its RediPlus, and I think there's one other firm that is prepared to do it as well. I believe it might be Morgan Stanley...I know with Goldman they used the Spear Leeds name for the clearing. But those are the only two that I'm aware of out of the major houses that are offering an ECN on the desk and also allowing sponsorships to SuperMontage anonymously.

Wheeler: RediPlus, ironically, is really more of an aggregator than access point. In fact, to get the most functionality on you default to Archipelago to get all those various order types, and you post on ARCA through . So there is some broker-sponsored access to , but we find it somewhat ironic that the cross functionality you get in those systems is through posting on Archipelago. 

InstitutionalInvestor.com: What are your thoughts on the level of program trading firms have been doing? 

Jenkins: I don't know if it would fall in line with program trading, but I guess just the competition for order flow. Most of the larger, top 10 brokers are creating algorithms which program traders use. We're going down the path of smaller programs - a few single stocks, but going through these algorithms that are either coming from their quant desk or their program desk. So that's something new. There's only so much of your business that's going to come in via program. It's got to be sent in multiple stocks. And if you're an active investor, you don't always have five things to do. Program trading will get up to a certain level. Last year it peaked at 25%, and that was probably more because we had a low of flows out and they were trying to get in the portfolio, rather than doing individual stocks. But now it's come back down again to about 20% of our business seems to be holding at that level. 

Cormack: The quantitative strategies that were employed by firms such as Automated Trading Desk and the have really just in the last 18 months continued to explode. more firms out there like that, firms you've never heard of, trading 5% of OTC volume every day. Some of one another or your orders or everyone else's orders. But there is a tremendous amount of liquidity out there within those groups. A lot of them are really making scratch or not even scratch on the trading and just living off liquidity rebates and tape revenue rebates. 

InstitutionalInvestor.com: Is that good or bad? 

Cormack: It's more liquidity out there, so when you want to add a bid or take an offering, it's there for you. Just theorizing some of the reasons why it's there, I think the market might be on a short-term basis, very short-term basis, too volatile. They're taking offerings and hitting bids when if they waited for five seconds or 15 seconds or 15 minutes, they could get those executions done at marginally better prices. That's what we're seeing more and more. I believe it might have topped out in the last few months. I can't imagine too many more people in that space. 

Connellan: On the active side, we are definitely seeing an increase in programs coming from our managers. Maybe because they are just modeling more and they're going to settle 2% of their portfolio or 5%, so they are capable of doing it now with the various order management systems that are out there. We are seeing an increase, but as Michael just mentioned, the space is getting crowded. Ironically enough, I think, on the sales side, that's not quite as profitable a business for them any more as it used to be, as I understand it right now, given the fact that there isn't too much in the syndicate. In the underwritings, the structure derivatives is the place for the most money, and then after that, the are the place where firms are deriving a lot of their profits - again, away from also the proprietary trading that they do. And the program trading has probably, to some degree, topped out. 

Wheeler: All I would add to that is there's definitely a transformation going on at the sellside level. The proprietary money is moving off the OTC desk, where it's almost pure agency now. The big high-dollar market makers are finding themselves out of work, or else they're moving over to the program desk. 

Connellan: They're even coming to the buyside, John. (Laughs)

Wheeler: Yes. We need them. I think all of these issues are interrelated, to try to bring this full circle, and if everyone is afraid to show their full order on the floor of the New York Stock Exchange, you can't get a fair execution by displaying size, so they break up their orders into smaller and smaller pieces, and they sell their business as a list, as a program trade, whether it's done as an agency trade or a principal trade. Back to the , again, if we fix a market structure, people can get the size done. By displaying size, I think a lot of this goes away. 

Buek: We have historically used program a lot. I think possibly why it's picking up is it enables you to be more efficient. Certainly the brokers have lost head count, so I'm sure they're pushing more to program trading when they have that ability to work a list with one person, rather than spreading it out. Also it keeps our anonymity, and you have so many more electronic venues to use now, especially . Seven years ago you had to separate the OTC orders and work mostly manually in tough negotiation with market makers. And now, it's predominantly electronic. The market makers used to have to have more relevant trading information when they were used more frequently, but since many are taking control and working their own orders the information flow has dried up as well.

InstitutionalInvestor.com: What about soft dollars? Does anyone think that at some point the SEC is going to tackle this, or will it just let it slide and continue going? 

Wheeler: I think the SEC is going to probably clamp down a little bit more on soft dollars. I think the SEC realizes that the should be able to pay for value-added research. My guess is they're going to push for more disclosure. I doubt that we will ever accept a model where research is paid for out of management fees rather than through the commission dollars borne by our shareholders. But I don't think you'll see 900 firms on a soft dollar list, like you can get right now. I think those days are gone.    

< Back to Press page >