Dealers will certainly be the big winners if Nasdaq's proposed limit-order facility is approved. They would regain priority over customer limit orders and establish a cheaper alternative to Instinet or Bloomberg Tradebook.
Institutional limit-order traders, however, would have a disincentive to use the system, and retail limit-order traders would find that many of their orders would go unfilled unless the market moves against them.
The limit-order facility, the centerpiece of a new integrated order-delivery and execution system, proposed and filed for approval with the Securities and Exchange Commission by the National Association of Securities Dealers, would create a central location where limit orders could be stored and prioritized by price and time. (The system was originally dubbed Next Nasdaq).
Like Instinet, the identity of the trader would not be revealed in the limit-order facility until the trade is executed. Institutions don't want to trade through dealers for fear of being front-run. On the other hand, dealers don't want their identities revealed until they have the chance to unwind positions they have taken.
At the moment, the main users of Instinet are Nasdaq dealers seeking anonymity they cannot obtain elsewhere. Last fall, another dealer market, the London Stock Exchange, instituted a limit-order facility. The evidence thus far indicates that the majority of the volume in that system is generated by dealers who have switched from other dealer-messaging systems.
The same should be more than true about the new Nasdaq facility. While dealers would find the new Nasdaq system a cheaper alternative to Instinet, institutions would probably remain on Instinet because of the Nasdaq system's live-quote rule.
That rule means that orders placed on the new system can't be canceled in the ten seconds following the order's entry. This would act as an effective deterrent to institutional usage of the system. (Note that market-makers' quotes on the current Nasdaq system are not subject to the ten-second rule.)
What about retail limit orders? Nasdaq is a fragmented market without system-wide priority rules. While a dealer cannot trade ahead of a customer limit order he or she holds, the dealer is free nonetheless to trade ahead of customer limit orders held by other dealers or electronic communications networks. Plans for the new Nasdaq facility call for price and time priority inside the facility. However, that is only envisaged for non-directed orders. The market practices of payment for order flow, internalization and preferencing would guarantee that little, if any, order flow would be non-directed. Therefore, the system-wide priority rules will rarely be invoked.
Market participants could direct orders to the new Nasdaq facility or to another dealer. In this case, dealers' best-execution obligations would require that they obtain the best price, not the first best price.
A dealer could direct his order to another dealer offering the same price as the new Nasdaq facility, and his customer would be no worse off than if the order was executed against limit orders in the new facility. Hence, given the practice of payment for order flow, there would be no incentive for market orders to be directed to Nasdaq's limit-order facility.
In contrast, the New York Stock Exchange and the American Stock Exchange provide sufficient incentives for a large amount of market orders to be directed to them for execution against their limit-order books.
Although brokers can direct market orders to one of the regional exchanges or to Nasdaq (in payment-for-order-flow or internalization arrangements), many do not direct orders this way. Price improvement is the reason the NYSE and the AMEX continue to control the lion's share of trading in their listed stocks.
Studies have shown that more than 25 percent of the orders sent to the NYSE execute inside the current quote. Brokers like to report price improvement to their customers, so they have an incentive to direct market orders to the national exchanges. This provides the flow of market orders necessary to make the NYSE and the AMEX limit-order books successful. (Regional exchanges provide some opportunity for price improvement, though a far greater opportunity exists on national exchanges).
Without system-wide time-priority rules and the chance for price improvement, there is no incentive to send Nasdaq market orders to the limit-order facility. And without the flow of market orders, Nasdaq retail limit-order traders quoting at the same price as dealers will see their orders go unexecuted. That is, unless the market moves against them, in which case dealers will move their quotes and limit-order traders will see their orders executed at the worst possible times.
This provides a disincentive for retail traders to submit limit orders. Without the flow of market orders to interact with, the proposed Nasdaq limit-order facility will fail.
At press time, Nasdaq announced an agreement in principal with OptiMark Technologies to allow access to OptiMark from Nasdaq on a commission basis. The preliminary plans are to allow Nasdaq's limit-order facility to take part in OptiMark matches.
Given the wide spreads on many Nasdaq stocks, it is highly probable that matches will typically occur inside the spread. Therefore, limit-order traders quoting at the same prices as dealers will be no better off under the agreement.
However, the agreement with OptiMark clearly shows that Nasdaq is beginning to realize the magnitude of revenues it can gain from commissions versus spreads. Nasdaq should consider system-wide priority rules for all trades. While the primary-priority rule should be based on price, the secondary rule does not have to be based on time.
For example, the secondary-priority rules on the Toronto Stock Exchange are designed so that each member with orders on the book is guaranteed at least a partial fill. An order exceeding the size of the first quote at a given price is shared among all quote participants on an equal-allocation basis, up to some level (2,000 shares). Any remaining amount is allocated on a prorata basis.
Nasdaq should remember the lesson it learned in 1983 when it fought hard to prevent contemporaneous trade reporting. Recall that then-NASD President Gordon Macklin met with the SEC six months after the NASD was forced to accept the rule for a small group of stocks, and asked that all of its stocks be subject to the rule.
It turns out that change was good for business, because volumes and profits increased after the rule change. I believe the same would be true if Nasdaq created a level playing field and allowed limit orders a chance to be executed.
Daniel G. Weaver is an associate finance professor at Baruch College in New York and an expert on market microstructure.
