A B A C U S    Issue 2.07 - July 1996

Market Remakers

By George Graham



In 1933, the town of Hartford, Connecticut, boasted a thriving stock exchange, a miniature Wall Street on Main Street. Then AT&T installed a long-distance telephone line to New York. Within two years, the Hartford stock exchange was extinct. Fields Wicker-Miurin, the North Carolina-born director of strategy and finance at the London Stock Exchange (LSE), likes to tell this tale to caution her colleagues about competition and changing technology.

On the face of it, the idea of London as a Hartford in the making is faintly ludicrous. The London Stock Exchange is, after all, the third largest equity market in the world - as measured by market capitalisation - after New York and Tokyo. Leading continental banks, such as Deutsche Bank of Germany and ABN Amro of the Netherlands, have recently made London the European headquarters for their trading activities.

But changing technology is turning stock exchanges arse over pin-striped tit. The automation of share trading is pushing exchanges everywhere to change the way they do business. At the same time, it is also eliminating many of the barriers of space and time that have heretofore kept exchanges from competing with one another. The combination of upheaval and new competition puts even the longest-established exchanges at risk; even the LSE faces no shortage of competitors, between continental exchanges and new British trading start-ups. And its march into the electronic future has been a stumbling one. The next year or so will be crucial for the LSE - a realisation all the more sobering given that, a mere decade ago, it was the LSE that threatened to decimate neighbouring exchanges.

Trading goes electronic

Back in 1986, the stock exchange reforms known as "Big Bang" moved the London market from the physical trading floor, where brokers had haggled over share prices, to the Stock Exchange Automated Quote system (Seaq), an electronic bulletin board where market-makers displayed the prices at which they were ready to buy or sell.

Seaq International lists the shares of overseas companies; for a while, it bit large chunks out of the business of the continental exchanges. But Big Bang was a digital revolution only partially accomplished. Although display of prices has moved onto an electronic system, actual trading is still carried out by negotiations on the telephone.

When the continental exchanges overhauled themselves in response to the LSE's challenge, they took progress to the next logical step: trading systems such as Paris's CAC-40 not only display prices, but also execute trades. Push a button and you can buy or sell at the price on the screen - without having to pick up the phone.

In 1993, the Stock Exchange's ability to keep up with the reviving competition was called into doubt by the collapse of Taurus, a system designed to automate the settlement of share deals - that is, the re-registration of shares and the exchange of payment that takes place after a deal has been struck. Infighting among vested City interests and incompetent IT management caused the project to collapse in disarray, with nearly £100 million pounds of partially finished work written off. In an implicit vote of little confidence, the Bank of England took over management of the replacement system, Crest, from the LSE.

So when Michael Lawrence, the LSE's chief executive, was abruptly defenestrated in January of 1996, both the cynics and the alarmists began to worry that things at the Exchange had gone from bad to worse. True, Lawrence did have the sort of abrupt personality that could lead to sudden departure. He had, for example, fired off lawsuits at some of the LSE's largest members. But Lawrence himself claimed to be the victim of a coup - one engineered by leading market-making firms such as BZW (the investment banking arm of Barclays Bank), which wanted to preserve their market share and the status quo against Lawrence's valiant attempts to drag the Exchange into the 21st century with a brand new electronic trading system.

In the aftermath of Lawrence's sacking, powerful forces in the Treasury, the Securities and Investments Board and the Office of Fair Trading realised that if the Stock Exchange rejected change outright it might cripple confidence in the market. So pressure was brought to bear and, by moving from general principles to details, the members of the exchange found some common ground. To understand the details, and the issues at stake, look first at London's competitors, and then at the changes which automation would bring to the most basic working of the market - the way it matches buyers with sellers.

Share boosters

The LSE's competitors come in several different flavours. New electronic exchanges promise to boost dramatically the efficiency of share trading. Tradepoint, for example, is a new London-based stock market that opened its network ports last year. Its computers accept a steady flow of requests to buy or sell, each at a given price. When the price of a buy order matches that of a sell order, the computer automatically strikes a deal and notifies buyer and seller accordingly. Because the flow of orders drives the pace of trading in the market, such systems are referred to as order-driven trading. The LSE, by contrast, is a quote-driven market. Instead of matching buyer directly to seller, most deals are channelled through market-makers, who buy shares with their own money and then re-sell them to buyers. The market-makers' quotes set the pace of trading.

It is not just stock exchanges that are using electronics to draw trading away from the LSE, though. BZW is an archetypal broker and market-maker in London, and hence one of the biggest traders on the LSE. But its fund-management subsidiary in California also acts as a private exchange, allowing the pension funds whose money it manages to "cross" shares with one another in a form of order matching. The volume of BZW's internal trading occasionally peaks at roughly half the New York Stock Exchange's turnover.

While few other established exchanges - like the New York Stock Exchange, Tokyo, Frankfurt, Paris, or Amsterdam - are as wired as either Tradepoint or BZW, they too could benefit from technology at London's expense. For communications technology now makes it possible to trade actively on any of these exchanges without setting foot outside of the square mile. And regulators are making it ever easier to do just that. The European Union's Investment Services Directive, or ISD, now allows an investment bank registered in one EU country to use that "passport" to trade in any other.

In the short term, the ISD will benefit London. Brokers like Union Bank of Switzerland can sign up as remote members of the Amsterdam or Stockholm stock exchanges while keeping all their dealing operations in London. But that also makes it easier for them to take their business elsewhere if the London exchange doesn't give them the services they need.

In response to these competitive threats, the London Stock Exchange has upgraded the platform on which Seaq operates with a £70 million development programme called Sequence. Sequence 6, the final stage, is due to be introduced on August 27th. Sequence 6 makes it technically feasible to operate a fully automated, order-driven trading system. But there are still big questions about just how automated London will allow its new electronic markets to become. Already, automated trading is to be limited to only big, high-volume shares. What other restrictions are put upon automated trading depends on the outcome of two very different arguments - one concerning costs and jobs, and the other concerning liquidity.

Automation could lead to large-scale redundancies among London's high-salaried dealers, and changing to a new trading system would also add to the pressure to pare down the Exchange's own cost structure. The Exchange still employed more than 1,000 people in 1995, and had operating costs of £177 million despite a serious attempt to cut staff and costs. Tradepoint, by contrast, has a budget of £6 million a year - although it lacks most of the Stock Exchange's regulatory responsibilities, and is of course much smaller.

For companies raising money on the London Exchange and the investors who provide it for them, however, the greatest savings from automation could come not so much through reducing the numbers of highly-paid traders they must support, but in changing the way prices are set on the Exchange. In London, commissions paid to brokers now average between 15 and 20 basis points - comparable to other international markets. This is dwarfed, though, by the spread between buying and selling prices. The "touch" - the gap between the best buying price and the best selling price to be found on the Seaq screen - may range from 0.35-0.45% for a liquid stock such as BP or HSBC Holdings, right up to 2% for the stocks at the bottom end of the FT-SE 100 index - and over 5% for less widely traded stocks. Even the keenest advocates of the current system acknowledge that this is wider than the average spread on electronic order books such as Paris's CAC-40 system.

Any threat to the spreads on the London Exchange, though, also throws open the second front of the present arguments over reform: liquidity. "It's irrelevant if you have the narrowest spreads and the lowest dealing costs, if you can't deal in size," says David Rough, who heads the investment management arm of UK insurer Legal & General. Although some investors focus on trading costs, many - possibly most - are at least as concerned with the ability to deal quickly, even in large quantities or in obscure shares.

London's large spread

A quantitative fund manager, whose job it is to track an index as closely as possible, is most interested in dealing at the middle market price and with the lowest possible transaction costs. Not surprisingly, index fund managers hate London. An active stock picker, by contrast, may be keener on the ability to secure an immediate price for a relatively little-known share. Any large institution, the typical holding of which may top £100 million, wants to be able to buy or sell without tipping its hand to the market and seeing the price race away from it.

A large spread provides substantial and obvious incentives for market-makers to keep a ready supply of shares for the market. And that is not to be scoffed at - particularly in tough times, when prices are moving fast and the difference between an instant deal and a delayed one can be hefty. In the 1987 stock market crash, for example, London was almost the only major market in the world to keep up trading without a break, even as the wave of sell orders hitting the market reached tsunami proportions.

However, many of the causes cited for London's reassuring record of maintaining liquidity even in difficult times are also linked to privileges granted to market-makers - which become irksome when times aren't so tough. The spread helps. Some also cite the exemption market-makers enjoy from government stamp duty, levied at 0.5% of the value of each share purchase; others attribute liquidity to market-makers' freedom to delay reporting some of their trades to the market. "Transparency, or rather the lack of it, is the driver," said one leading market-maker.

Deciding which of these privileges will be preserved in the face of automation is the most devilish question facing London as it gropes toward its future. In their defence, market-makers point out that few actually bear the full cost of this spread. Most institutions actually deal at better prices than appear on the screen, "inside the touch". In fact, allowing even small traders to avoid the high cost of dealing on the exchange is the key motivation for BZW to have its Trade system for matching client orders. Kleinwort Benson, too, has a proprietary order-routing system known as Best. Not only do the systems eliminate some market trades by matching buy and sell orders among customers, they also pool the remaining orders into larger deals - and so provide the bargaining clout to get well inside the touch.

The key compromise now being touted for the London Stock Exchange would allow small traders the option of matching their deals directly on the open market, via an automated, order-driven system. But there would be a parallel system for trading large blocks of shares, similar to New York's "upstairs" market. Here the big players could negotiate prices for blocks of shares too big for the market to absorb comfortably. The obvious and crucial question, however, concerns the relations between that upstairs room and the broad market below.

Market-makers prefer a loose relationship between upstairs and down. They do not want, for example, to have to fill small orders on the downstairs system at the same prices set in negotiated trading upstairs. Their critics counter that this will simply allow the big boys to keep the really good deals for themselves, and to gain an insider's edge on the market.

Seldom, even in the high-stakes world of the City, have the stakes in any deal been so high. Getting the right balance of risks and rewards is vital to London's future as a stock exchange. In the 18th century, when captains first began trading stakes in each others' cargoes in the coffee bars of St James's, anyone dissatisfied with the deal on offer could just go down the street to see if there was a more accommodating market at the next coffee bar. Today, however, traders don't even have to leave their desks to shop the world's exchanges. If the London Stock Exchange stumbles now, its custom will vanish in less time than it takes to type the name of the next exchange - wherever it may be.

George Graham is not the former Arsenal manager. He is a financial journalist at the Financial Times.


Thanks to the explosion of computer networks, new stock exchanges are popping up like mushrooms across cyberspace. Among London's erstwhile competitors, Tradepoint is the best established. It uses a proprietary network and Windows-based software to match buyers and sellers on-screen. So far, though, its turnover amounts to only 0.05% of the Stock Exchange. ESI, a Cambridge-based start-up which now provides financial information via the Internet, reckons the Net could expand the reach of stock markets to smaller companies - and it hopes to open an exchange later this year to do just that. Because, ESI argues, the costs of electronic trading are about a tenth those of the London exchange, it should enable deals to be done in companies a tenth the size by investors who need risk only a tenth as much on any given deal. For inspiration, both newcomers can look across the Atlantic to the Spring Street Brewery, a niche brewer founded by Wall Street dropout Andrew Klein. When it needed US$3.5 million to expand, it sidestepped financial markets altogether to raise money directly from investors who came to its Web site. So successful was this that Klein is now diversifying out of beer and back into share trading - this time with his own market, Wit Capital.