Thursday September 02, 2010




Research & Rankings

Latest Releases:
Latin America Latin America Research Team
BofA Merrill Lynch Global Research and J.P. Morgan share top honors in the 2010 ranking.
2010 All-Brazil Research Team
Itaú Securities leads annual ranking of Brazil's top equity and fixed-income analysts.
Coming in September:
Country Credit
Asia 100

 
Search by: Skip Navigation Links
RegionExpand Region
Skip Navigation Links
SubjectExpand Subject

SUBSCRIBE NOW to get instant, unrestricted access to current and archived research & rankings.

FREE TRIAL

Register today for a FREE 2 week trial including full online access and the latest issues of Institutional Investor magazine.

Inside the Machine: A Journey into the World of High-Frequency Trading

PRINT

By Michael Peltz
June 2010

Keywords: high-frequency trading, stock market, proprietary algorithms


Page 1 of 13

At 2:45p.m. on Thursday, May 6, George (Gus) Sauter received a frantic call from one of his traders to get in front of a Bloomberg terminal. The Dow Jones industrial average, already down 3.9 percent that day on fears about Greece, was in free fall. In just five minutes the index plunged 573 points. Less than two minutes later, the Dow had rocketed back up 543 points, going on to finish the day down 3.2 percent.

“It was just crazy,” Sauter, chief investment officer of mutual fund giant Vanguard Group, told me a few days later. “I had to go to our fixed-income building, about a five-minute walk from my office. By the time I got there, the market had rallied.”

high-frequency trading Crazy, indeed. The aptly named “flash crash” temporarily wiped out more than a half trillion dollars in equity value, shaking what little faith nervous investors had in U.S. markets. Shares of Dow component Procter & Gamble Co., the ultimate defensive blue-chip stock, dropped more than one third in a matter of minutes before recovering almost as quickly, all for no apparent reason. A few other large U.S. companies, including accounting firm Accenture, saw their stocks trade as low as a penny a share, only to close not far from where they had begun the day (nearly $42 a share in the case of Accenture) — again, on no news. By the time the dust settled, a whopping 19.3 billion shares had changed hands, more than twice the average daily U.S. equity market volume this year and the second-biggest trading day ever.

But for me, the single most amazing fact about the flash crash was that no one had a clue as to what had triggered it. Not that I should have been surprised, based on the conversation I’d had two days earlier with Mary Schapiro, chairman of the Securities and Exchange Commission. Schapiro, whose organization is charged with maintaining “fair and orderly” markets, explained to me how the SEC did a detailed study after the October 1987 crash to reconstruct what had happened. “We’ve lost some of the capacity to do that given the dramatic volumes of trading that exist today,” Schapiro said. “But we need to be able to do that to understand where are the vulnerabilities in our marketplace and what are the practices that have the potential to hurt investors and the marketplace in the long run.”

In 1987 the SEC had a much easier task because the vast majority of listed U.S. equities were traded in one place — on the floor of the New York Stock Exchange, where specialists employed by the Big Board’s member firms made a market based on an open-outcry auction system. Today, as a result of a series of regulatory changes designed to increase competition and make the market fairer for mom-and-pop investors, only about one quarter of all U.S. equity trading occurs through the now publicly held NYSE Euronext. And the majority of that trading is done electronically, either by the new NYSE floor specialists, called designated market makers, or on the fully automated NYSE Arca platform. The rest of the trading in U.S. equities is spread across a wide range of venues, including the three other major exchanges (Nasdaq Stock Market, BATS Exchange and Direct Edge) and dozens of broker-dealer-operated trading systems, electronic communications networks (ECNs) and dark pools, where buyers and sellers are matched up anonymously. 

1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13

Comments (3)

| Add Comments

roy Jun 21, 2010

In deciding whether this is good or bad, we have to realize the march of technology is inexorable, and this kind of rocket science will be seen as quaint in a few years. I also think that for regulators to have credibility (such as it is) they must support the rule of law consistently in environments that will continue to mutate. That means, don't try to ban this practice or point fingers and feign ignornace if it causes problems, rather insure that real competition is alive and well between practicioners. That will limit things going wrong.


Charles Jun 19, 2010

Robin: Hokey-schmokey, Batman, if 50% to 70% of all domestic equity trades are now controlled by super-sophisticated and super-speedy robots, and the robots are programmed to do nothing more nor less than squeeze out fractions of a penny on ask and bid prices, then what chance does Mr. and Mrs. John Littlevestor have in today's equity markets? Batman: Precious little, Boy Wonder, at least for now. Which is why the Littlevestors and millions of others are saying: No thanks, I'll take my marbles and go home, or look around for a more fair game to play in, one in which there is a fair to middling chance of actually winning. That's always been the American way. Level playing fields and all that. Robin: But Batman, if nobody is willing to play then the game can't go on. Batman: Precisely, young chum, precisely.


roy Jun 14, 2010

Naturally, everyone is worried about these people warping the market, and I'm sure they have economies of scale, but barriers to entry don't seem too high, because there are a million quant-savvy grad students out there looking for work. It's also possible that these firms get funding thrown at them by fat private equity, because they have a new idea, but it turns out to be harder to make money than was expected. Everyone thought the internet boom of the '90s would rewrite business, but it created hundreds of smart, hi-tech companies that washed out because margins were slim to non-existent. How many server farms were built in hopes of cornering the market on mail-order pet food? I hope the SEC is on top of potential shenanigans, of course, but I'm also happy to see the giant money-center banks have some competition. After all, corporate behemoths always seem to promote the lacrosse players, and give their long-haired geniuses short shrift.