Efforts to create a cheaper,pan-European stock exchange may be mired in chaos, but aggressive money managers and brokerages are finding ways to improve execution and rein in trading costs
By Justin Schack
November 2000
Institutional Investor Magazine
Cooperation sure doesn't come easily in Europe. Uniting much of the Continent under a single currency took years of political infighting and deal making, and the job remains only partially done. Not to mention how the euro itself has been battered.
Integrating the Old World's proudly independent bourses isn't proving any easier. Investors and major brokerage houses, eager to cut trading costs and improve efficiency, have been clamoring for one marketplace or common platform that would centralize stock-trading liquidity, reduce fees and make it easier and more likely for buyers and sellers to get the best prices. Nearly everyone agrees on this objective. But that's where the agreement ends.
The result has been a corporate farceur's dream: short-lived alliances, broken engagements, threats, insults and nonstop palaver. Most spectacularly, the ballyhooed merger between the London Stock Exchange and Frankfurt's Deutsche Börse collapsed in September, opening the door for Swedish trading system operator OM Gruppen and its astonishing hostile bid for the London exchange. Euronext, the combination of the French, Belgian and Dutch bourses that broke off from 1998's eight-exchange alliance, is maneuvering for a London connection, while Frankfurt and the U.S.'s Nasdaq may yet storm the City with bids of their own. LSE shareholders now obstinately insist that they can remain independent -- and create their own pan-European platform -- dimming the outlook for further consolidation.
"Everyone has been spinning their wheels for two years now. Not very much has been accomplished," says Richard Semark, head of trading for UBS Asset Management in London.
Despite all this, there's some welcome news. The cost of executing stock trades does appear to be declining on individual European exchanges. According to a global analysis conducted for Institutional Investor by Elkins/McSherry Co., a New York brokerage and trading consulting firm, execution costs fell last year on 11 of 15 major European markets, led by a 28 percent decline in Ireland. Four major markets in the region -- Germany, Italy, Portugal and Sweden -- became slightly more expensive.
Elkins/McSherry measures three cost components: commissions, fees and market impact -- essentially a measure of how much the size of institutional orders affects the price at which they are filled. If trades aren't handled properly, a negative price impact can cost investors far more than the commissions and fees levied by intermediaries. Market impact is arrived at by comparing the price at which trades are executed with the average of the daily opening, high, low and closing prices for each stock.
Commissions and fees, which have been under pressure for years, continued to decline on most European markets (Denmark and the U.K. proved to be exceptions). But the biggest driver of cost reduction was lower market impact. That fell by 78 percent in Austria, 56 percent in France and 34 percent in Switzerland, helping to bring overall costs in these markets down by 21 percent, 7 percent and 21 percent, respectively. In the U.K., market impact for sell orders fell by 27 percent, contributing to an overall cost reduction of 11 percent, even though commissions ticked up slightly. (U.K. buy orders are subject to a tax that inflates execution cost measurements.)
Europe's results stand in sharp contrast to surprising cost increases in the U.S., where institutions and brokerages have bellyached the loudest about the need to make European trading cheaper. The New York Stock Exchange, which barely edged out ParisBourse as the world's least expensive market for the second straight year, saw market impact increase by 8 percent, contributing to a 4 percent overall cost inflation. Nasdaq became 11 percent more expensive, fueled by a 14 percent uptick in market impact. Both markets implemented reforms in 1997 that helped reduce market impact in previous surveys. But unprecedented volatility, combined with unintended consequences of those very reforms, may have fueled last year's cost increases.
Overall, the costs of executing stock trades globally increased for the first time in the four years that Elkins/McSherry has tracked them for this magazine. The 2 percent increase, however, pales in comparison with the stunning 75 percent reduction from 1996 to 1998. Elkins/McSherry studied the global trading in 1999 of more than 150 large institutions. It measured activity in 42 countries by more than 1,500 brokerages and 800 investment managers. All told, approximately 1.2 million trades, representing more than 118 billion shares, were analyzed.
Successes notwithstanding, brokerages and money managers remain under pressure to reduce costs. Most significantly, Europe lacks a unified clearing and settlement platform; in the U.S. these operations are centralized under the Depository Trust Co. and the National Securities Clearing Co. Brokerages in Europe must therefore park expensive capital in several systems, raising the cost of cross-border trading ten to 20 times higher than it is across the Atlantic, industry professionals estimate. And although investors increasingly favor pan-European portfolios, the separate national exchanges force brokerages to staff duplicate operations in multiple cities and pass the costs on to clients. Moreover, as the amount of cash flowing into European stocks continues to soar, money managers must execute ever-larger trades, making it harder to control market impact.
"When you look at the largest investment management firms in Europe, the assets they have under management are increasing rapidly," says Marcus Hooper, head of equity and derivatives trading at Dresdner RCM Global Investors. "The problem is that if you double your order size, you're not just doubling your costs; they increase exponentially because of market impact."
Combining the exchanges and their back-office systems would go a long way toward addressing these problems. But as the past two years have shown, that's not going to be easy. The same nationalistic pride and institutional jealousies that stand in the way of exchange mergers also work to keep clearing and settlement entities from combining. "We're trying to achieve a single, central counterparty under one regulatory jurisdiction," says Pen Kent, a former Bank of England governor currently serving as chairman of the European Securities Forum, a group of brokerages pushing for a pan-European trade-processing infrastructure. "This will be difficult politically, because the French will want a bit, the Germans will want a bit and so on, and that's the reality of the EU."
Consolidation isn't the only way to rein in costs, of course. European portfolio managers have begun implementing so-called directed-brokerage and soft-dollar plans, which bundle research and execution services together to reduce costs and have been popular for a generation in the U.S. Furthermore, a growing number of institutions are weaning themselves from reliance on brokerages' capital to execute large orders, choosing instead to have brokers "work" some transactions over time to obtain better prices in the open market. Brokerages and managers are also turning to alternate methods to attain cheap executions, including electronic pan-European trading systems such as Tradepoint Stock Exchange and E-Crossnet. None has yet taken significant market share from London, Paris or Frankfurt, which, unlike U.S. markets, all adopted electronic order books of their own during the past decade. But investor awareness is growing.
"Two years ago, really, people looked pretty blankly at you," says Alisdair Haynes, CEO of ITG Europe, a subsidiary of New Yorkbased Investment Technology Group, which operates the Posit trade-crossing network. "Now I can go to any institution and say, 'I'd like to talk to you about execution costs,' and they really bite your arm off. There has been a very big sea change across Europe."
Driving this is a heightened cost-consciousness on the part of pension clients, who are using execution auditing services like those provided by Elkins/McSherry and rival firm Plexus Group to measure the efficiency of their investment managers and brokerages and to allocate their business accordingly. Investment managers increasingly demand that brokerages beat such execution benchmarks as the volume-weighted average price.
"I regard it as an extra auditor with which we can police the managers and hold them to account," says Joseph Barnes, head of investments at the British Coal and Mineworkers Pension, which began monitoring execution costs in 1997 and has shaken up its roster of managers as a result. "We get the feeling that often we're paying for things indirectly, because of the relationships between managers and brokers. This gives us more power to direct the execution decisions that typically have been made by the managers."
A number of factors underpin the changing atmosphere, ranging from technological advances to the fast-growing appetite for European equities generally. Net inflows into Europe-focused equity funds have nearly quadrupled during the past five years, from $1.4 billion in 1995 to $5.4 billion so far in 2000, according to AMG Data Services. As governments reform retirement schemes, pension managers long limited to such safe investments as debt instruments are warming to equities. And individual investors are gradually taking control of their savings and retirement planning, setting the stage for an explosion of retail investing like that in the U.S. during the past decade.
To cope with the influx of money, European asset management firms have begun structuring themselves more like their U.S. counterparts, with separate research, portfolio management and trading functions. Institutions buying and selling larger quantities of stock inevitably focus more closely on execution, and such specialization breeds greater sophistication. "As soon as you have someone whose job it is to get the best execution, the pressure is on that person to beat the benchmarks, or lose business," says Nigel Rowe, head of global equities at London investment bank Cazenove & Co., which ranks as the lowest-cost brokerage globally in this year's Elkins/Mcsherry survey.
Another driver of cost-consciousness is the growth of indexing among institutional portfolios. Market professionals estimate that some 20 percent of equity portfolios in the U.K. are passively managed, up from about 10 percent five years ago but still shy of the 25 to 30 percent level in the U.S. Passive equity management on the Continent has grown during that time from virtually nil to roughly 5 percent, and that's expected to grow further. With portfolios simply tracking indexes like the FTSE 100, transaction costs become an even bigger determinant of overall performance. "When you're indexing, the only thing that differentiates you from the crowd is your execution costs," says ITG Europe's Haynes.
In addition, U.S. institutions are funneling more money into Europe and taking their long-established focus on execution costs with them. "We're finding that the influx of U.S. money is causing the managers and brokers in Europe to pay far more attention to execution costs during the past year or so than they ever have before," says Elkins/McSherry chairman Richard McSherry. "If they don't, they know they're going to lose business."
With all these pressures on them, brokerages and asset managers alike are thus turning to little-practiced techniques to deliver cheaper trade executions. In London, for example, institutional brokerage was dominated for decades by large investment banks with huge balance sheets. These houses would commit capital to provide immediate executions for institutional block trades. Traders in the City estimate that some 90 percent of institutional orders were filled in this manner as recently as five years ago. Fund managers became spoiled, seeking immediacy without considering that better prices might have been had by asking brokerages to seek matching investor orders in the marketplace over a few minutes, hours or even days.
This way of doing business has slowly become the exception rather than the norm since the LSE introduced its stock exchange electronic trading service, or SETS, in 1997. Brokerages now have the ability to work institutional orders efficiently by breaking them up into pieces that are executed at the best possible price against other customer orders in SETS. Crossing orders -- finding an exact, natural match in the marketplace -- either through SETS or alternative systems like Posit or E-Crossnet is also becoming a sought-after and viable option. Buy-side traders (increasingly business-school-trained specialists rather than former floor brokers or more research-oriented portfolio managers) are warming to the ways that such strategies can reduce market impact costs. Today a slim majority of institutional orders are executed against other customer orders rather than against a big brokerage's balance sheet. Many institutions are even willing to pay higher commissions for crossed trades and other agency executions, as their overall transaction costs are still lower because market impact is limited.
"For the first year or so, it was a foreign concept for portfolio managers," Cazenove's Rowe says of the introduction of SETS. "They were trained to expect immediacy. Now, as dealing has increasingly become the job of a dedicated desk, the buy side is only asking for capital when it's absolutely needed." That has leveled the competitive playing field for smaller houses like Cazenove.
But order books sponsored by European exchanges, from the SETS system to Paris's CAC and Frankfurt's IBEX, do not offer complete anonymity to the parties involved in a trade. Institutions crave such secrecy. If their intention to buy or sell large blocks leaks to other market players, prices will respond accordingly, driving up the cost of doing the trade. Most exchange order books in Europe reveal the identity of the brokerage on each side of a transaction after it is executed. Though this isn't always a dead giveaway of a money manager's activity, it allows traders to make educated guesses based on their knowledge of client relationships.
This potential for information leakage, though not as severe as what can occur on a traditional open-outcry exchange floor, leaves the door open for electronic-matching systems like Tradepoint and E-Crossnet, which offer complete anonymity. Although they have yet to achieve the market share of their U.S. cousins (so-called electronic communications networks account for roughly one third of Nasdaq volume and a growing share of NYSE trades), institutions and brokerages alike are experimenting with any system that offers an advantage in the quest for lower costs. Indeed, electronic matching system operators Bridge Trading Co. and Instinet Corp. are ranked among the top 20 global brokerages for the first time in this year's survey. Bridge is second only to Cazenove globally, while 18th-ranked Instinet also placed among the top five brokerages in the Netherlands and Switzerland.
"I think electronic trading is one of the major factors in bringing down costs," says Dresdner RCM's Hooper. "The anonymity you get in the crossing systems is a real draw.
E-Crossnet is moving along, and Instinet is now well established in the U.K. and European markets. Definitely, the feeling is that there is value added."
UBS Asset Management, which owns Phillips & Drew, the top-ranked global investment manager for the third straight year, is turning to electronic matching systems to trim costs even further. E-Crossnet holds much promise in this regard, even though it's still struggling to establish a sufficient base of users and improve trade-matching rates: "We don't have a particularly good match rate at present," says UBS Asset Management's Semark. "But the expectation is that alternative systems will eventually take a bigger percentage of the marketplace."
Electronic upstarts are getting a boost from traditional bourses' lack of progress in merging to form a true pan-European exchange. After all, systems like Tradepoint and E-Crossnet are already pan-European. And they have powerful backers. Tradepoint is majority-owned by an Instinet-led consortium of U.S. firms. Morgan Stanley Dean Witter and OM Gruppen co-own Jiway, a proposed pan-European system for executing retail orders. And E-Crossnet, which would cut out brokerages entirely, is backed by the formidable duo of Merrill Lynch & Co. and Barclays Global Investors. European bourses, on the other hand, are viewed as national treasures. They are owned by local brokerages, national banks and sometimes governments, which often complicates strategic decision making. Merging them into one seamless unit will require either considerable diplomacy (which, in some cases, has already failed) or all-out war.
Consider the quandary of the LSE. In July 1999 it said it would convert from mutual to public ownership to enable it to combine with other exchanges. This April it said it would merge with the Deutsche Börse to create iX, a market for all European shares. But small British brokerages scuttled that plan, aiding OM Gruppen's hostile bid.
The LSE has managed to undercut OM Gruppen, pointing out numerous obstacles. One is the Swedish concern's majority ownership of Jiway, which would present a direct conflict with stewardship of the London exchange. Another is OM's convoluted ownership and governance structure, which includes a minority stake held by the Swedish government. OM would finance the takeover largely with its shares, which have fallen more than 20 percent in recent weeks. OM Gruppen raised its initial offer from about £750 million ($1.09 billion) to about £1 billion, but most observers doubt that it can prevail.
Publicly, London insists that it can go it alone. "We don't need to merge," says an exchange spokesman. "In spite of our withdrawal from iX, we remain intent on delivering a pan-European market. Our position here is strong: We remain the largest exchange in Europe, with a very deep pool of liquidity. But we're not going to rest on our laurels." Still, others see an organization in disarray. "The LSE at the moment is suffering from a crisis of self-confidence," says one person close to the 227-year-old bourse, who has knowledge of the ongoing talks among Europe's stock markets. "It's very vulnerable because it's a quoted company now. Anyone can come in and try to take you out."
The Deutsche Börse appears similarly isolated in the wake of iX's demise, particularly considering that it had twice come close to joining Paris under the Euronext flag before striking the ill-fated deal with London. Many fault the German government and its largest commercial banks, led by Deutsche Bank, which exert tremendous control over the bourse and shrunk from ceding that influence to foreign interests, for the lack of progress. "iX showed that it's hard for Germany to do mergers, except on their own terms, because of how strict their government is when it comes to these things," says one source. "But now they're in danger of being isolated and therefore are casting about for something to do rather urgently."
Two of those options appear unlikely to succeed, because of ill will carried over from aborted past deals. Frankfurt is said to be mulling over a hostile bid for the LSE, which London now is expected to fight with Churchillian resolve. Another possibility is a friendly combination with Euronext. Many on the Continent see this as particularly smart because of the trend toward Euroland investing. But officials at the Paris exchange, which leads Euronext, and those in Frankfurt don't see eye to eye after two hoped-for deals between them never materialized. "The two institutions are so distrustful of each other that there would have to be some kind of political intervention for anything to happen. There's quite a bit of bad blood there," says one well-placed source. "They had been in negotiations with one another for some time there and both thought the other side ratted on the deal, that they weren't negotiating in good faith. They would go to bed one night thinking they had made progress toward a deal and then wake up the next morning to find out the other side was lying through its teeth."
Euronext is also considering a bid for the LSE, but most think it will prefer instead to compete with London for the near future. Euronext is well positioned among the pan-European contenders, thanks to a single clearing and settlement platform, which Germany and London have yet to assemble.
Nasdaq, which would have allied itself with Frankfurt and London under iX, is seen by some as a potential savior of the LSE. Nasdaq chairman and CEO Frank Zarb, who had been expected to announce his resignation long ago but recently had his contract extended by one year, has long envisioned seamless trading of all the world's stocks on a single platform. A merger with the LSE would be a bold, if risky, step in that direction. (None of the markets would comment about any negotiations or potential combinations.)
One possible trump card that frustrated brokerages may play: focus on the back office, build their own trade-processing system and cut out the current competitors. European Securities Forum member firms are considering drawing up plans for such a system and opening up the bidding for its construction to outside contractors like banks or technology firms. "We could give the winner our business and back it with our money. That's a real possibility," says ESF chairman Kent.
Despite all the frustration, all the fits and starts, European stock investors are at least paying lower transaction costs than ever. Surprisingly, the same cannot be said for those playing in the U.S. market. More disturbingly, these markets grew more expensive largely because of greater market impact -- the very segment of execution costs targeted by the reforms implemented some three years ago.
Why? Some observers point to unintended consequences of the regulations. A federal investigation of Nasdaq dealers revealed that many kept customer limit orders to themselves so that the spread between what they could buy and sell a given stock for would remain artificially wide, boosting their profits in the process. Subsequently, regulators mandated new order-handling rules that forced brokerages to display widely any customer limit orders that would improve the best quoted buy and sell prices. Then, in the summer of 1997, U.S. markets narrowed their minimum price variation from one eighth to one sixteenth of a dollar.
The cumulative effect of both reforms, seen later in 1997 and throughout 1998, was the narrowing of spreads -- a major component of market-impact costs. The order-handling rules, mandated by the Securities and Exchange Commission, also gave a huge boost to ECNs, which technology-poor market makers used for displaying customer limit orders widely.
Even as the new rules were heralded, some on Wall Street and in academia warned that the new environment would shrink dealers' profits so much that many would withdraw from market making, thereby reducing liquidity for investors. Others criticized the fragmentation of trading interest among so many ECNs, charging that this, too, would compromise the ability of institutional investors to buy and sell large quantities of stock. Sure enough, major dealers like Merrill Lynch, PaineWebber and Salomon Smith Barney slashed hundreds of less liquid Nasdaq stocks from their market-making rosters, and ECNs soon claimed one third of Nasdaq volume.
Though there is no evidence proving a direct link between any of these developments and higher market-impact costs, some traders are more than willing to question federal regulators' wisdom.
"I don't think everyone should take all this wonderful comfort in narrowed spreads," says Andrew Brooks, head of trading at Baltimore, Marylandbased mutual fund complex T. Rowe Price Associates. "Market makers pulled back, and people are afraid to trade. Intraday volatility is through the roof. I mean, some of these stocks have 4-point spreads. How can you possibly expect to get a good execution in that type of environment? The fact is our executions are getting worse, and we represent individual investors. All the people who are trumpeting narrowed spreads and the glorious effect they've had on the market are really missing the boat. Competition is a wonderful thing, but somehow we've got to bring things back, so people can access pools of liquidity."
Regardless of the cause, U.S. markets should worry about becoming costlier places to do business. Competition today may be NYSE versus Nasdaq or LSE versus Euronext, but in the not-too-distant future, it will be global -- and the lowest-cost service provider will win. As more pension and mutual fund managers discover the importance of execution costs, brokerages and exchanges will no longer be able to rely on investors' naivete about the true cost of trading stocks.
Calculating trading costs
The Elkins/McSherry Co. analysis measures total trading cost, which consists of execution commissions and fees added to a calculation of trading effectiveness called "market impact." Market impact is the difference between the price at which a stock trade is executed and the average of that stock's high, low, opening and closing prices during the day.
Elkins/McSherry ranks countries and exchanges by aggregating the total trading costs of all of their stocks. Individual brokerage house and money manager rankings measure the amount by which the firm's average total trading cost beats that of the market or country as a whole.
