The core question is simple: What's the best way to structure the mechanism by which people trade stocks?
Investigators at the SEC and elsewhere are still looking into the flash crash of May 6, when the Dow Jones Industrial Average dropped nearly 1,000 points and then quickly recovered.
There are several theories about the causes of the crash — automated trading gone awry, options-trading strategies that triggered violent equity moves, inter-market communication glitches and plain-old market manipulation. But no one has quite figured it out.
Two weeks ago, BlackRock Inc. released a study it conducted among 380 financial advisers about the crash, and the majority said that market structure issues — notably an overreliance on computer systems and high-frequency trading — probably were the main drivers.
Ah, market structure. It's an issue that academics, regulators and executives at the markets themselves have been arguing about for decades. The core question is simple: What's the best way to structure the mechanism by which people trade stocks?
The answers are complex because there are many ways to interpret “best.”
Does “best” mean cheapest, fastest or something else? And who is it “best” for — individuals, institutions or traders, the owners of the marketplaces or their users?
Most people, of course, don't give a hoot about and don't understand market structure, unless a scandal erupts or stock prices, for no real obvious reason, plunge by 1,000 points.
Contrary to the assumptions of some of my colleagues, I, in fact, was not there to observe the market structure debate under the buttonwood tree when traders formed the New York Stock Exchange to swap Revolutionary War bonds. Nor did I witness the Pecora hearings about the failings of the NYSE in the crash of 1929.
But I do remember the debate over the creation of a national market system in the 1970s. Congress mandated the creation of such a system as part of the Securities Acts Amendments of 1975, which ended fixed commissions. Of course, Congress never said what a national market system was or how we would know when we actually had one. Its main purpose, it seemed, was to end the dominance and quasi-monopoly of the market makers known as specialists on the New York Stock Exchange, where I worked at the time.
While Nasdaq had been created in 1971, it was not much more than an electronic bulletin board linking securities dealers. As the tech boom took off, Nasdaq's multiple-dealer market turned into a serious rival to the NYSE's auction market, which again raised market structure issues. Which is better — several market makers offering competing bids and offers, or a central market maker where most buy and sell orders come together producing a price that reflects aggregate supply and demand and can be satisfied without dealer intervention?
A tough question, and a long one at that. The thing is, this fundamental question was never really answered. Nasdaq suffered a price collusion scandal and turned itself into a true electronic marketplace, while the NYSE effectively bought out the specialists and floor members and became the owner of its own trading mechanism.
So how do these structures work? That's like asking how Google finds what you're searching for. Who knows?
Back in the olden days when I worked at the NYSE, I would take visitors to the floor and they would ooh and aah at the bustle. They didn't have a clue about why the floor brokers were running around or what the activity was all about, nor did they much care.
Today, the brains of the New York Stock Exchange are probably in some non-descript server farm near the New Jersey Turnpike or the New Delhi-Mumbai Expressway. Nobody much cares now either — until the server crashes.
