Back when "online stockbroker" meant "a guy waiting for a drink at Harry's," investors didn't have to care about where their stocks were traded. Today's investors don't have that luxury. Online brokers have empowered them with more information and faster trades. But a proliferation of markets and electronic trading sites means a brokerage's decisions on where to send a stock order can determine how well and quickly your trade gets executed--or even if it's executed at all.
All those exchanges, market-makers, and electronic communications networks (ECNs) that match buy and sell orders automatically are in hot competition, even paying brokers to send orders their way. Wherever they send trades, brokers have a legal obligation to get "best execution"--the best possible price in the shortest reasonable time (table, page 184). But the Securities & Exchange Commission has found some brokers put payments they get from dealers ahead of their duties to customers. "Some firms appear to be allowing payment for order flow to affect which markets they send their orders to--at the expense of quality executions" warns SEC Chairman Arthur Levitt Jr. SEC examiners found these firms--which the commission won't identify--weren't monitoring the quality of executions, let alone striving for the best.
The SEC promises to lean on such negligent brokers. But that's not enough. The average investor now has no way to tell which brokers will deliver quality trades. What's needed is sunshine: clear, understandable, and uniform data that the public and the press can use to compare the quality of stock trades from broker to broker and market to market.
Brokers have that data. All the big houses and most smaller ones pay consultants to audit exchanges and dealers who execute their trades. They just don't share the numbers with customers. If investors had the data, brokers would be forced to compete on the prices they obtain for customers as well as the speed of their trades. That competition would benefit individual investors and U.S. markets alike.
Best execution is hard to pin down. Some traders favor speed above all; others would wait seconds or even minutes while a broker shops an order around to improve the price by a few cents per share. But real money is at stake: On a 1,000-share buy order, the chance to save a "teenie"--a sixteenth of a dollar--can be worth $62.50. Even at a midprice broker like Charles Schwab (SCH), that price break would cover the commission for the trade--and the next one. "With today's low commissions, investors need to start focusing on other transaction costs," says John Markese, president of the American Association of Individual Investors.
The biggest of those costs is the spread--the difference between what dealers pay for stocks they buy and charge for stocks they sell. Getting a price inside the spread can slash an investor's overhead. But price improvement is a game of chance. Dealers can't improve prices on stocks that trade with tiny spreads. Even among eligible trades, only a fraction get improvement: around 50% on the New York Stock Exchange floor, about 30% on NYSE-listed shares traded away from the floor, and 20% of Nasdaq trades.
Given those odds, most brokers think retail investors define good execution as a speedy trade. At online broker DLJdirect, surveys show that "customers want speed," says Michael Hogan, general counsel, "and so that's what we focus on delivering." Speed--and low costs--are the appeal of electronic market centers, like ECNs and the proposed small-order automated execution system that the NYSE is considering. Trouble is, the SEC has found that many firms had no evidence that speed was their customers' top priority. Nor were brokers auditing order routing to ensure that they were using the fastest and most efficient dealers. "If they aren't asking what customers want or what their dealers are delivering, it leaves the strong impression they're routing simply on the basis of the payments they get for order flow," says an SEC official.
STANDARDS. Those payments aren't inherently bad. Order flow from investors has value. Some brokers keep that value to themselves by buying market-making dealers--Schwab, for example, owns Mayer & Schweitzer--or by investing in ECNs. Other brokers sell orders for a penny or two per share, using the proceeds to cut the price that investors focus on--commissions. Academics who've studied the issue say that, of all these "preferencing" arrangements, payment for order flow is the easiest to monitor and the one most likely to benefit customers.
But brokers who shop hard can get both payments and price improvement for their customers. Bernard L. Madoff Investment Securities, the dealer that pioneered order-buying for NYSE-listed stocks, promises to improve prices on a high percentage of its trades. So it makes little sense to just denounce order-flow payments, as some critics do. To shop for the best broker, investors need to know more than they can learn from current disclosure rules, which require brokers to tell customers who ask whether their orders were sold.
What's needed is a standard, industrywide scorecard that focuses on brokers' performance. It should balance the complexity of order-routing with the clarity investors need. The key figures: What percentage of a broker's orders get price improvement, and the average savings, plus the time it takes to get those trades. The data would need to be segmented--breaking out NYSE and Nasdaq stocks and orders of different sizes--but brokers shouldn't be allowed to bury investors in numbing detail. The goal: to let reviewers and customers determine which brokers' execution arrangements work best for different types of investors.
Ideally, the securities industry will develop standards for the data brokers should release. If it won't, the SEC has the clout to force disclosure. Investors shouldn't have to wonder whether today's increasingly byzantine markets are costing them the trades that they deserve.