The buyside wants an ECN. The traders on the floor want an electronic replica of the floor. The New York Stock Exchange is trying to please both. Will the hybrid' fly? Will it pass muster with the Securities and Exchange Commission? And, if so, will the buyside use it?
It's been two years since the Big Board first filed its plans with the SEC to incorporate more automatic executions into its mostly manual trading environment. Since then, the exchange has filed seven amendments to the proposal (with another on the way) and acquired an electronic competitor. Contentious issues remain and SEC approval does not appear imminent.
"The plan is not in trouble," an NYSE official told Traders Magazine. "We expected we would have had full approval of hybrid by the end of last year, but because of all of the changes here and at the SEC that was not possible."
Indeed, the regulator recently gave the New York the go-ahead to phase in some of its new rules and technology on a pilot basis.
Dual Pressure
Driving hybrid are regulatory pressures and customer complaints. The buyside wants more automatic executions and so does the SEC. The SEC's new trade-through rule will go into effect this June, effectively forcing the Big Board to automate or risk having its quotes ignored.
Its response is the semi-manual, semi-electronic hybrid' marketplace. Trades can occur within the Big Board's computer systems or on its floor. The proposal is a compromise that tries to balance the demands of those on and off the floor.
Institutions will get more automatic executions, but perhaps not as many as with an ECN. Specialists and brokers will retain much control over the trading process, but will likely participate in fewer trades.
The New York's transformation is not revolutionary, but evolutionary. The Big Board already offers automatic executions via its Direct+ system. In operation since 2000, Direct+ handles 12 percent of total NYSE volume.
Direct+, though, only accepts orders of 1,099 shares or less and requires waits of 30 seconds between orders. The elimination of these restrictions is expected to lead to a surge in electronic trading. That could jeopardize the need for a trading floor.
Yet, while opening up Direct+ is a dramatic step, the Big Board is not going fully electronic. The combination of Direct+ and the specialist's display book will not function as an in-house ECN.
Two major design elements will tie the trading on the book to the trading on the floor, giving floor traders more control over Direct+ than they have today. The end result is a transformation of Direct+ from a mini-ECN into an extension of the floor.
First, and foremost, specialists and floor brokers are being given special privileges. They will be able to use special reserve files and order types, for instance, not permitted to outsiders.
Second, the exchange is keeping (but reworking) the circuit breaker that now shuts down Direct+ in the event a stock runs too far, too fast. Once tripped, orders fall into the hands of the specialist.
The New York controls about 75 percent of the trading in its listed securities. But for the exchange to remain relevant in a post trade-through world-where all markets will likely offer instant fills-the buyside must believe they will get a fair shake with Direct+.
So far, they are not happy. "The proposal falls short of providing an effective trading system that will encourage investors to place orders on the Exchange," the Investment Company Institute (ICI) told the SEC last year.
"I used to give hybrid a B+," said Jamie Selway, a principal at institutional brokerage White Cap Trading and the former chief economist at Archipelago. "Now, it's sort of C+/B-. I like the ECN elements, but have concerns with the special privileges [of the floor traders]."
By and large, money managers would prefer to see something along the lines of an ECN. That's not happening though. The combination of the specialist's display book and Direct+ does not an ECN make.
Yes, as with an ECN, a high-bid or low-offer limit order placed on the book is subject to an immediate execution. And, yes, as with an ECN, a trader can sweep the book to take liquidity.
But while the NYSE's set-up looks like an ECN on the surface, a closer examination reveals a more circumscribed environment.
ECNs, for instance, operate under strict time priority-orders are filled in the order in which they are received.
With hybrid, the NYSE has hardwired its time-honored concept of parity. That means first-come, first-served, but not necessarily second-come, second-served.
The first order that posts on the book at a given price gets filled first. But the second order (and the third, etc.) must share any incoming contra size with every other order at that price.
That's how orders-those from the book and those from floor brokers-are handled in the crowd today. "Parity encourages crowd participation in the price discovery process," the exchange maintains.
That concept is outdated, according to one critic. George Rutherfurd, a Chicago-based consultant to the buyside, told the SEC that parity is "simply a euphemism for no price/time priority.'" Rutherfurd, believes orders on the book should trade under strict time priority.
The parity problem becomes worse for would-be liquidity providers when floor broker interests are factored in. Those limit orders that came in second or third, for instance, may have no standing at all if there are floor broker orders on the book.
For the purpose of parity, all public limit orders at a given price are consolidated into one big order. Each individual floor broker order, however, is treated separately.
The upshot is that the first public limit order at a price will get filled, but the rest will have to wait until all the floor broker orders at that price are filled.
Liquidity providers might also have to kiss their standing on the book goodbye in the event Direct+ is halted. Under the proposal, auto-ex will stop temporarily if a stock runs too far, too fast.
In the Crowd
When stock prices (quotes or trades) reach certain levels known as liquidity replenishment points (LRPs), trading moves instantly to the crowd. Traders with priority orders on the book could lose their standing and end up on parity with the crowd.
Some broker-dealers consider this unfair, believing those orders deserve automatic executions. The Securities Industry Association (SIA) has complained to the SEC "an order that may have caused the price of a security to move to a LRP would be denied priority coming out of the LRP."
Parity for Specialists
Floor brokers may have some parity problems of their own. Many are upset with an exchange proposal to change its rules and put certain specialist orders on parity with their own.
Today, specialists are allowed to trade on parity with floor brokers in the crowd if they are liquidating or decreasing a position. That helps them shed unwanted positions at the end of the day that they took on in the performance of their stabilization duties.
Specialists, however, cannot trade for their dealer accounts on parity with brokers if they are establishing or increasing a position.
In the first iteration of the hybrid plan, this parity rule was extended to the electronic world. Specialists' displaying limit orders had to yield to both public and floor broker orders.
Under pressure from the specialists, though, that changed. Specialists can now trade on parity with broker orders on the book and in the crowd regardless of their positions. They still must yield to public orders.
Some floor brokers are seething. "Lifting that restriction would tilt the playing field in favor of the specialists and to the detriment of the floor brokers and, more importantly, to the detriment of the floor brokers' customers," Warren Meyers, president of the newly formed Independent Broker Action Committee (IBAC), recently told the SEC.
IBAC is calling on the SEC to reject the parity changes and the hybrid proposal. Others are too.
The ICI informed the SEC: "The Institute opposes eliminating this restriction. Placing specialists trading for their proprietary account on parity with investor orders misaligns the interests of participants on the exchange."
Go-alongs
The NYSE defends its position. "In a hybrid environment involving more automatic executions availability," the NYSE told the SEC, "the specialist's position should not dictate whether he or she is on parity with the crowd."
At least one floor broker agrees with the exchange. Robert McCooey, chief executive of the Griswold Company, notes floor brokers will often allow specialists to trade alongside, or at parity with, their go-along orders.
"Many customers like company," McCooey said. "They don't want to be 100 percent of the volume at any one price. So, in the absence of other institutional buyers or sellers at that price, the specialist may provide that company."
The concept of parity is not exclusive to the NYSE and is part of the rules at other stock and options exchanges. But, according to buyside consultant Rutherfurd: "It is common knowledge the SEC staff has for years disdained the NYSE's "parity" rule, "but has felt powerless because the rule was of longstanding."
Under hybrid, limit-order traders will also have to contend with competition from another source: the "price improvers."
Specialists and floor brokers are being given special trading privileges that would allow them to trade against incoming orders before they have a chance to match up with limit orders on the book.
By bidding higher or offering less than the sitting limit orders, the floor traders will earn the right to capture the trade.
That's known as price improvement and is considered the lynchpin of the price discovery process at the New York. The incoming order gets traded at a price better than expected. That sitting limit order however gets nothing. To the owner of that order, he has been pennied' or robbed.
Under the proposal, NYSE specialists will receive an API (application programming interface) from the exchange that allows them to write algorithms to monitor incoming orders and quote or trade electronically on behalf of their dealer accounts.
The floor brokers will not get APIs, but the ability to use so-called "discretionary orders." Electronic discretionary orders have long been supported by electronic markets and are popular with traders.
In the Weeds
With them, traders can display at one price but also maintain a hidden, more aggressive price. If a suitable contra order whizzes onto the book, the hidden discretionary order can grab it. Another trader's best bid or offer gets nothing. In industry jargon, the orders allow a trader to "hide in the weeds."
That means traders can get pennied on ECNs just as they might under hybrid. The difference is that, under hybrid, they will not have the ability to hide in the weeds themselves without a floor broker's assistance.
The New York has promised to give floor brokers such discretionary orders, but has been slow to publish details. That has IBAC, the floor brokers group, worried.
It is concerned that brokers are losing the arms race of trading functionality to the specialists. It wants the means to counteract the sophisticated algorithms being given to the market makers.
IBAC has petitioned the SEC to withhold approval of hybrid until the NYSE provides some details about the discretionary orders.
"No hybrid market proposal should be considered until the NYSE supplements its existing submissions," the floor broker group told the SEC.
The ICI is concerned about the specialists' special privileges as well.
"Allowing specialists to electronically interact with incoming orders in this manner and, in effect, step ahead of investor orders," the ICI complained to the SEC, "runs counter to the NYSE's goal of providing incentive to investors to place orders on the exchange."
Bad Things'
Specialists naturally disagree. "The specialist has to be able to react to orders that come in," Michael LaBranche, chairman and chief executive of the Big Board's largest specialist firm, said recently, "otherwise he can't prevent bad things from happening."
By bad things,' LaBranche is referring to excess volatility. He added: "You do need to see the orders before they are executed and before they are auto-ex in some situations. We're talking about very, very small amounts of time. We still have an affirmative obligation going forward."
The exchange, in defending the algorithms, told the SEC that without them "specialists will have distinct difficulty fulfilling their obligations to maintain liquid and continuous two-sided markets."
Robert McSweeney, a senior NYSE executive, has also noted that without the ability to price improve, the market might bifurcate.
"It would be extremely inviting for broker-dealers with significant retail flow to internalize that flow," McSweeney said at a recent industry conference. "Those customers would not have the ability to trade at prices better than the market's best quotes."
McSweeney added that other customers would not have the opportunity to interact with the internalized flow because it would be removed from the price-discovery process.
The specialist's ability to price improve is subject to certain guidelines. His algorithm can only react to one incoming order at a time and must abide by certain penny-per-share constraints.
NYSE Relents
When the parameters were first proposed, specialists complained loudly to the NYSE that the restrictions would put them out of business, sources told Traders Magazine.
The NYSE relented and re-proposed more liberal guidelines. The rules now give the specialist more opportunities to trade and remove some of the risk involved.
In the first hybrid draft, the specialist could not improve upon a quote if the spread was only 2 cents; now he can with a penny.
In the beginning, the specialist had to top the quote by 4 cents or 5 cents if the spread was greater than 10 cents; now he only has to price improve by 3 cents.
The specialist still can only react to one incoming order at a time. "We gave in on that one a little bit," noted Tom Schaefer a specialist with Van Der Moolen Specialists at a recent industry event. "We're not going to be able to see five different orders as we do now. Only one."
Both the specialists' algorithms and the floor brokers' discretionary orders are intended to work in conjunction with hidden reserve files.
These "agency interest files" and "specialist interest files" are similar to the reserve orders available to traders in the Nasdaq world. They permit a trader to display a token amount while hiding the bulk of his order. That lets him get size done without showing his hand.
With hybrid, these strategies are only available to floor brokers and specialists. Both groups maintain they need them if they are to adapt to an auto-ex world.
Predetermine
Now, specialists and floor brokers have some time to react to incoming orders. With auto-ex, there will be no time. So, floor traders will have to pre-program their trading interests to anticipate changes in prices and size.
"With hybrid we will have to predetermine our interest to buy or sell at a certain price," McCooey, the floor broker, said. "We need reserve orders to make sure if we get swept that we will be available without hitting an LRP."
Van der Moolen's Schaefer agrees. "We will have to anticipate where we want to be," he said, "and get our interest in prior to receiving orders. That is the biggest challenge. We will have to be smarter and not so reactionary."
Hidden reserve orders rank lower than displayed orders in the competition for a fill. But they allow floor traders to participate in sweeps in theoretically greater amounts than outsiders. Hidden orders are likely to be larger than displayed orders due to the lower risk involved.
Floor brokers are likely to use their so-called agency interest files, or e-quotes, to expose their go-along,' or participatory, orders to potential sweeps.
Rutherfurd and others, however, believe money managers should be able to input their own hidden go-along orders rather than pay a floor broker for the service.
"This forced intermediation is expensive, unnecessary, anti-competitive, and clearly intended to maintain an artificially protected niche for floor brokers," Rutherfurd told the SEC.
Despite the relative handicaps, limit-order traders face with hybrid, the platform does offer at least one advantage over ECNs. An order priced away from the inside could receive a better than expected fill in the event of a sweep.
Limit orders placed on ECNs trade at no better than their limits. Under the hybrid plan, those placed on the specialist's display book at prices away from the inside may fill at the so-called "clean-up price."
In a sweep, it is expected that, most of the time, an incoming order will trade against all available limit orders until filled or the supply of shares is exhausted.
The sweep order fills at two prices: the top of book and the clean-up price, or the price of the last order necessary to complete the trade.
So, if the best bid is $20.00 and the last bid necessary to fill an incoming sell sweep is $19.90, all those limit orders in between execute at $19.90.
A buy limit order on the book priced at $19.95 then gets done at $19.90, or 5 cents per share less than expected. In other words, it gets price improved.
Once again, the New York is attempting to integrate its ideas about price improvement into the automated world.
Of course, the trader who sent in the sweep order might not appreciate the outcome. In the ECN world, he fills at every price level he sweeps. With hybrid, he may be forced to price improve some standing limit orders, resulting in a worse fill than hoped for. Some believe that the clean-up price arrangement could deter traders from sweeping the book.
The biggest concerns of traders contemplating a sweep are the circuit breakers, or LRPs. If a stock moves too far, too fast, Direct+ shuts down and trading reverts to manual mode.
The LRP's purpose is to curb price volatility and preserve the exchange's reputation as a stable trading environment. Companies that list their shares on the Big Board abhor volatility, NYSE execs maintain.
"They focus on the volatility of their stock," Roger Burkhardt, the NYSE's chief technology officer, said at an industry conference last year, "which they believe relates to their ability to raise capital at attractive rates."
Burkhardt said he expects the LRPs to activate no more than 1 percent of the time.
Outside traders generally oppose the LRPs, preferring zero halts or at least more liberal guidelines. Barring that, they want to be able to predict when an LRP is likely to activate so that they can program their order routing systems.
The LRP Enigma
Bear Stearns, for example, is on record requesting the New York Stock Exchange supply stocks' short-term volatility data so it can better understand how often LRPs will be reached. Timber Hill, a leading derivatives trading house, has also asked for trading data.
Traders can use immediate-or-cancel (IOC) orders when sweeping. Under the proposal, the IOC will cancel if an LRP is triggered.
White Cap's Selway doesn't like the LRPs, but is not overly concerned. He believes the providers of order routing systems will incorporate functionality to predict the likelihood of an order triggering an LRP.
"The aggregators, whether it's direct market access or algorithms, provide useful management of market data," Selway noted.
The prospect of such innovation encourages Selway to believe the buyside will find value in hybrid. The trader also expects the New York to hold its own in the new marketplace despite its inability to offer unfettered auto-ex.
Survival
As an analogy, he offers the position of Instinet after the NASD's order handling rules spurred the creation of a group of new ECNs in January 1997.
"There was a lot of new whiz-bang functionality that was better than the clunky old Instinet screens and book," Selway says.
"But Instinet was largest. They didn't die. Eventually they did shed some share. But I think the same will hold for the New York. The transfer will happen a lot slower than people think."
The Arca Card
If the hybrid concept fails with the consumer, the New York Stock Exchange has a fallback: Archipelago.
The all-electronic marketplace bought by the Big Board last year already trades listed names and that volume is expected to grow. If traders decide that the ECN-like hybrid is not good enough, Archipelago, the real thing, could benefit.
Some even question the need for the hybrid now. "The design is a compromise between an ECN and a floor," Jamie Selway, a principal with White Cap Trading and the former chief economist at Archipelago, said. "You have to wonder about the need for it. Why not just run Arca and the floor side by side?"
The New York Stock Exchange had no comment.
