How High-Speed Traders Are Like Fish
At precisely 4:00:00 p.m. on Dec. 5, closing time for the Nasdaq stock exchange, a company called Ulta Salon Cosmetics & Fragrance announced worse-than-expected earnings. By 4:00:01 p.m., in the one second Nasdaq required to calculate the final stock price, high-speed traders with access to direct news feeds had already knocked it down by about 3 percent.
The case of Ulta demonstrates why timely information is precious, and how trading on it is getting ever faster. Computers process news so quickly that they can move stock prices before exchanges even have time to calculate them. Technology companies promise traders the power to place more than 20,000 orders a second over a single high-speed data link.
So what, if anything, does all this speed do for the ordinary investor? As I have written, it's certainly capable of creating new kinds of market instabilities. To be fair, it also has its benefits.
To operate smoothly, markets must respond quickly to information, which itself is arriving at an increasingly fast pace. If they don't, stock prices won't reflect the available information, and regular investors will run a larger risk of buying or selling at the wrong price. This function of price discovery is one of the fundamental reasons that markets are considered superior to other forms of economic organization.
New research by Austin Gerig, a physicist and quantitative finance expert at Oxford University, suggests that high-frequency trading improves markets' price discovery function. It does so by ensuring that the prices of assets that ought to move in sync actually do so.
Suppose, for example, that bad weather increases the cost of raw materials needed for making soft drinks. Coca-Cola stock immediately falls, but the shares of less prominent companies in the soft drink industry don't move as quickly. The market takes time to digest the information, draw out its implications and properly synchronize prices.
Using detailed trading data from Nasdaq, Gerig finds that high-frequency traders enforce and significantly speed up the synchronization process. Back in 2000, it took several minutes for a price change in one security to be followed by a similar price change in other related securities. Now, with the advent of high-frequency trading, it happens in less than 10 seconds.
This ought to be comforting news for average investors who act over much slower times and don't have vast resources to gather and process information so quickly. As the Ulta Salon example suggests, computing and communications technology has made slower investors vulnerable to those with better information. But without HFT, these vulnerabilities would be worse. The high-frequency traders, in seeking their own profits, end up helping ordinary investors to make better decisions more quickly. In the Coke-Pepsi example, the slower investor would pay the right price, rather than too much to someone who knew that Pepsi stock would soon fall as it followed Coke's.
In this sense, Gerig's study backs up what some prominent investors have been saying for years. In a letter to the Securities and Exchange Commission in 2010, Vanguard Chief Executive Officer John Bogle defended high-frequency trading, saying that it "enables investors to get a fair price across market centers" and helped achieve a "'knitting' together of the marketplace" that is broadly beneficial to long-term investors.
There is also an interesting scientific angle to Gerig's work. His point is that synchronization of prices aids decision-making for the average investor. Scientists have come around to a similar view of synchronized behavior in biology -- in flocks of birds or schools of fish, for example. We usually think there has to be a trade-off between speed and accuracy, but flocks or schools use synchronization to make decisions both more quickly and more accurately. A school of fish, for example, pools resources by scanning its environment with many eyes all at once, and synchronization enables all the fish to react quickly to any threat.
None of this means that high-frequency trading is an unmitigated good. We know that such traders tend to bail out of markets in times of trouble, withdrawing their orders to buy and sell when they are most needed. The Financial Industry Regulatory Authority is wisely aiming to develop new rules to stop questionable practices such as "momentum ignition," in which computer programs try to trick one another into generating price movements from which they can profit.
But in one particular way, at least, all the high-speed activity does seem to help the average investor. The truth is rarely simple. There's no harm in admitting that.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
(Mark Buchanan, a physicist and the author of “Forecast: What Physics, Meteorology and the Natural Sciences Can Teach Us About Economics,” is a Bloomberg View columnist.)
--Editors: Mark Whitehouse, Stacey Shick.
To contact the author on this story:
Mark Buchanan at email@example.com
To contact the editor on this story:
Mark Whitehouse at firstname.lastname@example.org