Lupien got it wrong.
Will Merrin get it right?
Seth Merrin, the technology pioneer widely credited with launching the order management system industry, is now vowing to do what others tried but failed at: eliminate market impact on block trades.
Merrin's new venture, Liquidnet, hopes to provide access in one gigantic pool of orders, the liquidity in the OMSs at the top buy-side firms.
The plan: Traders using Liquidnet will be electronically notified when the system has the other sides of orders on their blotters. The traders would then have the option of negotiating the trades with their counterparties.
"We are creating a country club kind of network," said Merrin, chief executive of Liquidnet, a 30-person operation based in New York. "All the members have the same problem-huge orders they must execute."
"The market impact is killing them because right now brokers are the only way they know to match up with other institutions," he added. "Market impact is an enormous problem."
Merrin had a hit ten years ago when he introduced the Merrin Financial Trading Platform (now owned by The MacGregor Group). Since then the installed base of these order-tracking and routing devices has ballooned to some 700. All told, they account for an estimated one billion shares in order flow each day, according to the Tower Group, a securities consulting group in Needham, Mass.
Traders agree that market impact, or the effect of a large trade on the price of a stock, is a costly overhead. But while they like the Liquidnet concept, some foresee practical problems in using it. "They say it's a passive system, but in practice the trader is going to have to make adjustments everyday on his OMS, or else he is going to run into problems," said one trader familiar with Liquidnet.
The Liquidnet scheme has three parts. First, it plugs into the OMSs of the largest trading desks over the Internet and amalgamates all the orders. Second, it displays to traders select contra-orders to those in their OMSs. Finally, it lets traders negotiate price and quantity among themselves via text chat. No sales traders are involved so no information is leaked, Merrin says. That eliminates the possibility of market impact, he says.
Unlike other trading systems, there is no inputting or monitoring of orders. Finding the other side is a completely passive process on Liquidnet. Liquidity is brought to the trader. The trader does not have to bring liquidity to the system.
Traders never know with whom they are dealing. They also do not know the total amount of the stock available for sale. They only know there is enough to at least match a predetermined minimum portion of their order.
Minimum quantities negotiated are based on parameters established by traders when they first connect to the system. A trader might specify he will only accept contras with at least 25 percent of his order. That way a buyer of one million shares, for example, would be dealing with a seller of at least 250,000 shares, if at all. (A million-share buyer would normally not want to deal with a 5,000-share seller.)
Liquidnet is only courting the largest institutions with OMSs. It says 134 have indicated an interest. Launching in December, it will not charge "members" a license fee for the technology. As a broker dealer and Securities and Exchange Commission-registered alternative trading system it will charge a pennies per-share fee on each transaction.
On its board of advisors are the head traders of Aim Advisors, Scudder Kemper Investments, Putnam Investments and T. Rowe Price.
"It holds great promise," said Leo Smith, Putnam's head trader. "It takes the promise of OptiMark and makes it easier. Ultimately it will be successful."
Smith notes that too many traders settle for the VWAP, or the volume-weighted average price of a stock over the course of the day, when transacting. "This will force people to make decisions. We're supposed to be traders, not VWAPers."
Like all other traders interviewed for this story he does not believe Liquidnet will completely replace the sales trader. "I look at it as another avenue to source liquidity," he said. "It's not totally doing away with anything."
Liquidnet is not the first piece of technology to tackle market impact. Three systems aimed at block-traders have come and (mostly) gone over the years.
OptiMark Technologies, founded by Bill Lupien, was a multi-million dollar failure now partly on the auction block. @Harborside, an indications service, was yanked from the market by parent Jefferies after only six months in service. (Creator Richard Holway says it will be available again soon following some enhancements.) The Arizona Stock Exchange, a call auction system near death earlier this year, won a stay of execution when Goldman Sachs took a stake. All three suffered from a lack of liquidity.
Traders give various reasons for the failures including questionable anonymity, difficulty of use, time-consumption, and too many "nothing-dones." Merrin says Liquidnet avoids the first three problems and shifts the burden of the fourth to the trader.
Nothing-dones aren't likely to completely disappear, however, because a trader is under no obligation to trade. "They're not committed orders," Merrin acknowledged. "They're only indications really."
"Indications" in the strictest sense are those sellside-to-buyside e-mails suggesting trades. They have come in for criticism as brokers posting indications are often just "fishing," or trying to gauge the market with no intention to follow through. Thomson Financial's AutEx and Bridge are the two biggest operators of IOI networks.
The slippery nature of the "orders" concerned Kevin Cronin, head trader at AIM, at first. "That was one of the things I worked hard on," said Cronin, an adviser to Liquidnet. "We want to make sure the people in this system are people who really want to trade."
The system will monitor and score traders' accounts based on how often they negotiate. Other traders will be able to use those scores to determine the seriousness of a contra. If a trader frequently refuses to negotiate his rights will be revoked, according to Cronin.
"These are not just brokers fishing," he said. "[With Liquidnet] you can start the negotiation process with a guy because he says he's looking for exactly what you have."
Another trader predicts the rating will hurt "good" traders. They will get black marks from contras for not matching orders they consider unexecutable, but in the blotter anyway. He cites three examples. First, a user transacts over the phone while his order is still in Liquidnet. An angry contra watches a print go up and gives the trader a black mark.
Second, the trader is unwilling to negotiate an order on the blotter at current prices. Again, the contra complains. Third, the trader refuses to negotiate an order filled but not yet deleted from the OMS. Some buy-side shops don't clear out their OMSs until the end of the day when they allocate.
"Guys are going to have to start marking the orders they want to go into the system and those they don't," the trader said. "Once you get to that point what happens to the promised ease of use'?"
On the flip side, Liquidnet appears to solve problems traders had with OptiMark and @Harborside. Because it allows them to see some size they are not flying blind as with OptiMark, or "OptiDark" as it was uncharitably called. "OptiMark never really helped us to understand supply and demand," Cronin said.
Because Liquidnet discloses no names it says it guarantees anonymity. Traders were never comfortable with the two sales traders in the middle of an @Harborside transaction.
Middlemen are considered the root of the market impact problem. Sales traders at brokerages leak information that moves prices. Even traders at successful systems like Instinet and Posit are said to leak information. (Officials at Instinet and Posit deny this.)
Liquidnet is, of course, a middleman. And it plans to amass a huge amount of very valuable trading information. Merrin claims it's impenetrable. "Nobody within Liquidnet can see what is going on," he said. The startup hired accountants Ernst & Young to audit its processes and procedures to certify Merrin's claim.
Some traders are still not convinced of the utility of electronic middlemen. Brian Pears, head trader at mid-sized Wells Capital Management, is skeptical that bypassing the sales trader will necessarily result in better performance numbers.
"I'm not sure how much demand there really is for institutions only to be talking to each other," Pears said. "We all talk like that, but I think the buyside relies on the sellside for a lot more than they think they do. Liquidnet is an interesting idea, but it won't revolutionize the trading community. I don't think it will ever gain critical mass to the point where people say this is the way to trade."
Pears says a good sales trader can save him from making a bad trade. If he buys the first 100,000 shares from a seller of 2,000,000, the price he pays may look ridiculous once the remaining 1.9 million hit the Street. His sales trader can forewarn him that the other side may have more to do.
"That's what we would lose in negotiating over text chat with another buy-side trader," he said. "The sales trader is incented to protect their client somewhat."
Pears says he is willing to try the service, but it won't become a regular part of his routine unless it proves itself with liquidity.
Merrin acknowledges his system won't completely replace sales traders, but says the liquidity will be there:
"If we start off on Day One with the 100 largest institutions on the Street," he said, "we're going to have more market information in our system than the exchanges."
