“Flash Boys,” Michael Lewiss blockbuster expose of fast trading in the stock market, has hit a huge nerve. This is not surprising: the pace with which trading occurs in todays electronic markets is mind boggling, the cost of attaining todays speeds is staggering, and the associated thought that the markets are rigged is a real attention grabber.
The staggering costs should be controlled, and there is a way to do it: Treat all orders that arrive in a brief interval — perhaps a second or half-second — as if they had arrived at the same point in time, batch them together, and clear them at the same time and price.
Traders have traditionally sought to sniff out each others intensions, and the quest for greater speed has always been integral to the stock market. The question is, should anything be done about it and, if so, what? While proposed solutions include tweaking market structure with speed bumps, special fees, and so forth, we need to dig deeper. We need to question the very need for a hyper-continuous market that allows a trade to be made at any instant that a buy and sell order meet in price.
Remember the saying, it happened in the blink of an eye? A person’s blink takes roughly 350 milliseconds, each of which is one thousandth of a second. In todays markets, orders and trades whip around far faster than that. Microseconds (a millionth of a second) and nanoseconds (a billionth) are on the scene. Now, as long as the pressure to be fast does not get out of control, there is nothing wrong with speed. Participants do not want to wait an hour, a minute, or even less to get an order into the market, realize a trade, and receive a report. In and of itself, does a millisecond have any intrinsic value?
Yes, it certainly does for a horse race where a winner is declared and the payoff for being first is big. Our markets share this property of a horse race. As long as technology enables time to be measured in ever smaller fractions of a second, the race will continue on regardless of how hyper-continuous the continuous market becomes until we have reached the speed of light.
What Michael Lewis has portrayed in “Flash Boys” comes across as dramatic because a millisecond world is dramatic. But, 200 years ago, a carrier pigeon was dramatic. In a book I published in 1988, I wrote that, after Napoleon was defeated at Waterloo on June 18, 1815, A carrier pigeon took news of the British victory to Nathan Rothschild in London. In a single day, Rothschild reaped a fortune by buying shares from uninformed, and quite frightened traders (he was also credited with having saved the London Stock Exchange). Rothschilds profit was not due to the news. It was due to his having received the news first.
Should aspersions be cast on a pigeon? I do not think so, but neither should trading be thought of as a horse race. Fortunately, an alternative to a hyper-continuous market exists: let trading progress, not in continuous time, but via a rapid succession of discrete points of time. That is, impose a minimum time tick on trading (perhaps every second or half-second). Referring to the procedure as staccato trading, I proposed this with my colleague Liuren Wu in The Journal of Portfolio Management, Spring, 2013. With staccato trading, all orders that arrive within the same brief interval (e.g.,a second) are treated as if they had arrived at the same time.
The process of batching orders together and clearing them at the same time and price is commonly referred to as a call auction. The call auction way of handling orders puts investors and traders on a more level field, and it enables more reasonable prices to be produced. Importantly, it mitigates any feeling that, in some way, the market is rigged. And the rapid, staccato succession of calls would deliver another benefit of enormous importance: it would curtail the extremely costly arms race for ever greater speed.
For the broad community of investors, spending exorbitant sums to save a couple of milliseconds makes no economic sense at all. To deal with such miniscule intervals of time, the time clock should be used differently: call auctions should be held with, for instance, one-second or half-second frequency, while the brief time between calls should be used for order assembly and trade determination.
Disentangling order assembly from trade determination will sharpen price discovery. And perhaps the intensely competitive players on Wall Street and beyond will realize the collective advantage of slowing the market down a bit so that competition based on speed might be brought under better control.
Robert A. Schwartz is the Marvin M. Speiser Professor of Finance at the Zicklin School of Business, Baruch College, CUNY.