When the International Securities Exchange (ISE) began trading options contracts on May 26, 2000, David Krell, its president and CEO, knew all eyes in the financial markets were watching his every move. Not only was the ISE the first entirely electronic options exchange in the United States, but it was also the first exchange, period, to be approved by the U.S. Securities and Exchange Commission (SEC) in 27 years. Its mere existence raised the question of whether options trading might become entirely electronic in the next few years, replacing the venerable "open outcry" system of floor trading as it has in markets around the globe.
The prospect stirred up the options industry well in advance of the ISE's launch. When plans for the all-electronic market were announced a year and a half ago, the power of the digital age had yet to energise the options industry--with broadly available instant execution and large-scale anonymous trading--the way it had the stock markets. Plus, fees were still considered high at the floor-based exchanges, and competition among them had been lax. The backers of the ISE saw a huge opportunity to introduce a lower-cost, screen-based market to both whisk away market share and grow with the boom in the options business in general.
Indeed, the market for options has exploded in the past few years, right along with the roller-coaster ride in the stock markets. It's easy to see why: Options contracts allow the purchaser to buy or sell a stock (known as calls and puts, respectively) at a given price, no matter what the value of the actual stock might be at the moment. The difference is the profit the purchaser can claim, often in cash. That contract, paid for by a premium like insurance, allows the investor to hedge an actual investment or just speculate on the movements of a stock. The recent gyrations of the stock markets have caused interest in options to spike. Investors have used them as a way of protecting against losses and sudden fluctuations, with total annual contracts traded jumping from 295 million in 1996 to 508 million last year.
Yet for all its potential, the launch of the ISE was underwhelming. The exchange began trading with just three of the 600 options contracts it expects to list by year-end. After six weeks, it offered 34 options, with total daily volume of 15,000 to 20,000 contracts. By comparison, the total U.S. options market (made up of approximately 2,500 stock options) currently trades in the range of 2 million to 3 million contracts per day. Clearly, there's a long road in front of the ISE.
But as Daniel Friel, CIO of the ISE, points out, the slow start is all part of the plan. "We started small deliberately in order to work out any small problems we might have and allow our market makers to get comfortable with the exchange gradually," he says.
The goal is for the ISE to roll out more quickly this fall and trade all 600 of its intended listings by the end of the year. But even then, the race will have just begun. It's an analogy Krell understands well. "This is a marathon, not a sprint. We don't see the success of the ISE coming in two months or six months. This will be a long struggle, and that's what we're prepared for," he says.
A VULNERABLE FIELD
As is true for any marketplace, the key to long-term success will be not in the number of products it offers but in the number of customers it attracts and the amount of business they want to do. In the options business today, most customers turn to the Chicago Board Options Exchange (CBOE) or the American Stock Exchange, followed by the Pacific Exchange and the Philadelphia Stock Exchange. Luring a steady flow of orders from brokerages and other trading firms onto the ISE will be a tough task. At the least, the ISE will need to prove itself a flawless vehicle for trading from a technical and operations standpoint. Then the exchange will need to impress options traders with compelling reasons to use the system--faster executions, lower trading costs and complete anonymity. But even winning that battle may not win the war for market share.
"It's a bit of a chicken-and-egg situation for the ISE," says James Marks, an e-commerce analyst at Credit Suisse First Boston Corp. in Philadelphia. "To get order flow, they need liquidity--willing buyers and sellers--but to get liquidity they need order flow. Better, cheaper, faster won't mean much if they don't get the critical mass of order flow they need to keep their market makers and the brokerages happy."
When plans for the exchange were first announced, the four floor-based options exchanges in the United States were vulnerable, making the possibility that the ISE could win over substantial market share more than an empty threat. First, all four had been slow to fully embrace new technologies and to allow fully automated trading of larger and more complex orders--something brokers value because it allows them to anonymously yet quickly trade larger orders on the market by breaking the orders up to avoid shifting the price through their concentrated demand, known in the business as "moving the market."
"On the floor exchanges, the floor brokers always can see the orders coming--they get 'second look,' which enables them to raise the price when they think a large order is coming through, particularly from an institutional investor, like a mutual fund or pension fund manager normally represented by the same brokerage firm," says Krell. "On our exchange, market makers won't be able to tell if they're dealing with Fidelity Investments (Inc.) or Mrs. Jones. It's more fair to everyone."
Moreover, despite SEC encouragement since 1989 to develop a more competitive market system, the traditional exchanges have been reluctant to compete, preferring instead to trade their self-developed rosters of options, with few multiple listings. The options business was steadily growing, and each exchange was able to list contracts in new companies on a regular basis without overlap. In addition, the technology available prior to the late 1990s made multiple listings difficult to manage in a real-time environment. "It's not that exchanges didn't want to compete head-on--there was no real need or even ability to do so," says Ed Joyce, president and chief operating officer of the CBOE.
That all changed in November 1998 when William Porter, founder of E-Trade Securities, announced that he and a collection of partners would launch the ISE. Not only did Porter and company have the cash and savvy to pull off such a feat, but the new exchange would list the most actively traded options of all four existing exchanges, establishing direct competition with the incumbents.
Over the next year and a half, the four exchanges announced new technologies and trading rules to shore up their franchises and new product lines and market structure changes. Plus, they all began more actively developing competitive business in one another's more-active options and slashing fees for customers.
The biggest of the four, the CBOE, has moved aggressively to parry the ISE's challenge. The CBOE has long attempted to keep up with available technologies to help its members do more business more efficiently, such as its adoption of an automated routing system, handheld wireless trading computers for traders and a computer-based trading platform for smaller orders, now up to 50 contracts. But critics point out that the CBOE's systems have neither worked as well as planned nor created substantially better functionality.
"The biggest problem for the CBOE is that it is conflicted," says Larry Tabb, group director for the securities and investment practice at the TowerGroup, a finance technology consulting company in Needham, Mass. "It's owned by its members, and the members are biased toward floor trading and against automatic execution, because that is the business they know, and it's how they make the most money. Its technology initiatives have had problems and have always taken longer than expected as a result."
But the CBOE won't be caught without an answer to this criticism for long. According to Joyce, his exchange is developing CBOE Direct, a screen-based trading system that will be available by the fall for both early morning and after-session trading. "We'll expand it as we see fit," says Joyce. "If the market wants to trade electronically, they will be able to do it with us."
Even so, electronic exchange analysts believe the ISE should maintain a head start for some time. "They're doing all the right things to attract and excite the liquidity they need to get their market up and running, and they have a good system and market structure," says Marks. "They just need to make sure the technology works flawlessly."
FROM DREAM- TO DEAL-MAKER
For Friel, ensuring that the ise's technology design would result in a bomb-proof system presented a laundry list of imperatives and performance goals. Most important, the trading system could never present a trader with a pause, let alone a roadblock, to interfere with quickly and securely executing a trade. To ensure that, it needed to be capable of handling tens of thousands of bids and offers per second, routing them within the market and executing them if possible. If the exchange's market makers couldn't execute an order because it was outside of their given quotes for a contract, the exchange system had to be able to automatically send it to another exchange--an "away" market--if market makers there would execute at that price. In addition, the exchange's records would need to be fully traceable and auditable for surveillance purposes. To top it off, the system could never fail--it had to be fully redundant, from the core trading system to the gateway servers sitting on brokerage trading floors around the world to the fiber-optic lines linking it all together.
Today, Friel believes, such a system is in place. Creating it, however, was tougher than expected. ISE turned to OM Technology, a Stockholm, Sweden-based software company that has developed the trading systems for 13 other fully electronic markets, including the Australian Stock Exchange, the Hong Kong Futures Exchange, the Milan Stock Exchange and OM Stockholm Exchange, most of which include options trading as part of their business.
Yet key software customisation issues gave the development team pause. First, the system had to be able to not only handle thousands of competitive quotes broadcast through the system every second, but it also had to integrate the best bids and offers on the other exchanges so that member firms could keep their prices up-to-date. That data flows from the Options Prices Reporting Authority (OPRA) at a rate of 3,500 quotes per second, and that number is expected to hit 12,000 per second by year-end and upward of 50,000 per second over the next few years. Plus, the best bids and offers on the ISE needed to be routed to OPRA to keep other markets appraised of its market movements.
"All of the other options exchanges OM had developed were the only markets in their system. To incorporate that external data and make it possible to send an order to one of those 'away' markets was a big challenge," says Friel.
Second, the ISE had to be able to scale up the application and hardware quickly to keep up with the explosion of quoting and trading activity expected over the next few years. A key part of that was ensuring that the hardware and network choices worked well with OM's system. Because all of OM's code was written for and running on a Compaq Computer Corp. OpenVMS platform, using Alpha servers, Friel wasn't about to try something new just to be different.
The final system design allows the ISE to upgrade the 17-node core exchange system to add power or memory where and when it's needed. And the network is redundant at every step: A completely separate backup market system in Jersey City, N.J., runs in sync with the New York City-based main system. The two are linked by a pair of OC3 lines, and all the network market participants are linked via two independent frame-relay networks--private T1 lines in the New York area and redundant fiber-optic networks through WorldCom Inc. for participants elsewhere. At each location, the market participants use two Compaq Proliant servers to follow the market and execute trades. The two market system sites have their own generators in case power is lost.
WILL THE TRADERS COME?
Krell and friel knew there'd be no Field of Dreams magic awaiting their trading platform once it was up and running. The key to making the exchange successful would be selling the benefits of the system to large trading firms and brokerages, encouraging them to embrace it and, in turn, attracting other customer orders. Accordingly, the core partners in the ISE, called Adirondack Trading Partners, invited large securities firms to become "market makers"--that is, willing buyers and sellers of certain options, no matter what the market might do. To lower the risk of taking on such responsibilities, the market makers would be granted pools of 60 contracts to trade, to balance the risks of one contract against another. For each "bin" of 60 options, there would be one primary market maker (PMM) and 10 competitive market makers, all offering buy and sell quotes at all times. Thus far, PMMs include Banc of America Securities LLC, Bear Specialist/ Hunter Specialists, Deutsche Bank AG Securities, the Hull Trading Co. owned by Goldman, Sachs & Co., KBC Financial Products USA, Knight Financial Products LLC and Morgan Stanley Dean Witter & Co. --all heavy hitters in the financial world.
But why would unaffiliated brokers and other traders decide to send order flow to the ISE when they could easily get their business done through existing channels? First, the firms making up Adirondack Trading Partners, including Ameritrade Inc., E-Trade Group Inc., Knight Financial Products and Herzog, Heine Geduld (recently bought by Merrill Lynch & Co.), have plenty of their own order flow from customers and see the exchange as a new tool to trade faster and more cheaply than is possible on the traditional exchanges. (Conflicts of interest are avoided because the market-maker and brokerage operations of these companies are separate, and the ISE operates anonymously.) Transactions on the ISE are free to both brokerage and customer. Market makers are charged 24 cents per contract, and the turnaround time on an order is less than one second. This compares with a fee structure that was about 50 percent higher at the other exchanges early this year. Since then, the CBOE, for one, has lowered its market-maker fees to match the ISE and has waived all customer and floor broker fees as well. The other exchanges are expected to follow suit. Meanwhile, floor orders can take anywhere from 15 seconds to several minutes, depending on the order and market conditions at the time. The ISE's increased speed and reduced cost should allow market makers to better control their bids and offers, thus profiting more with every trade. This should also allow them to offer more competitive prices than those at the existing exchanges, which should draw more orders to the ISE as a whole.
BIGGER EVERY DAY
This combination of benefits has Krell and the founders of ISE convinced that the exchange will not only capture a respectable share of the options markets over the next few years, but it will also help create growth in the market. "Just as we've seen in the stock markets, as trading becomes cheaper, faster, more automated and more private, volumes will increase. We've seen that the last year or two with the advent of multiple listing and lower fees, and we expect it to continue," he says. As for an overall share of the market, in the medium term, says Krell, "we'd be happy with 5 [percent] to 10 percent of the market."
What the ISE team doesn't predict, at least publicly, is a mass exodus from the other exchanges to its network. Trying to soften the competitive hype swirling around the ISE's debut, Krell says, "I don't expect a huge shift of market activity to us from the CBOE or the others. Just as there are clear benefits to a screen-based market, there are great benefits to floor trading." In an electronic setting, for example, it's difficult to execute complex transactions or to pursue a complex, multistock hedging strategy. Krell believes trading floors and floor brokers will continue to do strong business and won't be entirely replaced by computers--the ISE's or anyone else's--anytime soon. "The other exchanges shouldn't worry about us too much," he says.
Nevertheless, if the ISE works as well as intended, it will continue to put downward pressure on fees and upward pressure on technology budgets at all the other exchanges, giving them plenty to worry about for years to come.
Ian Springsteel is a freelance writer based in Somerville, Mass. He can be reached via e-mail at email@example.com.
A FASTER ROUTE
Speed is one big advantage to the ISE's electronic exchange On the International Securities Exchange:
1. John Trader pulls up his e-brokerage account, which is wired to place orders on the ISE preferentially, and places his order to buy 60 contracts on shares of America Online Inc.
2. The order zips through the e-brokerage's server gateway and hits the ISE main system.
3. The trading system routes the order to the "bin" where AOL options are traded.
4. Trader's order is split by the rules of the system among three market makers offering the best price on the 60 puts, each offering the put in blocks of 20 contracts. None of them knew Trader's name or his brokerage's, or that he wanted a total of 60 contracts instead of 20.
5. A confirmation is routed back to Trader through the e-brokerage's site.
Time elapsed: About 1 second
0.2 seconds for the transaction, about 1 second for the entire process Through the Chicago Board Options Exchange:
1. John Trader places his order through his account with Bull & Bear, which has a preference for sending its orders to the CBOE.
2. The Bull & Bear system routes the order to the CBOE's Order Routing System (ORS) feed.
3. The ORS determines that the order is too large to flow into the CBOE's retail automatic execution system, which is limited to 50 contracts or less. So it is instead routed to one of the floor brokers responsible for AOL contracts in the AOL pit.
4. The order pops up on the floor broker's screen, visible to the crowd of 12 market makers and other brokers trading in AOL options around him. The floor broker announces he has an order for 60 puts at the market.
5. One market maker immediately indicates interest in 30 contracts at a slightly better price than the last trade, and another market maker follows suit. The floor broker indicates the deal is done and presses a few keys.
6. The confirmation is routed electronically back to Trader via Bull & Bear's website.
Time elapsed: 1.5 seconds
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.