How Speed Traders Are Changing Wall Street

It may surprise you to learn that most of the stock trades in the U.S. are no longer being made by human beings, but by robot computers capable of buying and selling thousands of different securities in the time it takes you to blink an eye.

These supercomputers - which actually decide which stocks to buy and sell - are operating on highly secret instructions programmed into them by math wizards who may or may not know anything about the value of the companies that are being traded.

It's known as "high frequency trading," a phenomenon that's swept over much of Wall Street in the past few years and played a supporting role in the mini market crash last spring that saw the Dow Jones Industrial Average plunge 600 points in 15 minutes.

Most people outside of the industry know very little, if anything, about it. But the Securities and Exchange Commission and members of Congress have begun asking some tough questions about its usefulness, potential dangers, and suspicions that some people may be using computers to manipulate the market.

Today it is still the public façade of Wall Street, and a television backdrop for reporters relaying financial news. But less than 30 percent of the trading is conducted there now, and the specialists and the noise of the floor is being replaced by the speed and quiet efficiency of computers, and the action has moved elsewhere.

There are now more than 80 alternative trading systems around the country, plus two brand new electronic stock exchanges which most of you have probably never heard of: BATS and Direct Edge.

They're owned by the big banks and by high frequency trading firms, and neither of them would give "60 Minutes" an interview or let us inside to film their operations, but they trade more than a billion shares a day at blinding speed, and most of those bets are being made by machines.

The players range from firms like Goldman Sachs, Barclays, Credit-Suisse and Morgan Stanley to hedge funds and smaller operations like Tradeworx, which is the only high frequency trading firm that would talk to us or let us in.

It's run by Manoj Narang and a small group of mathematicians and scientists called "quants," which is short for quantitative analysts. Their high speed computers trade 40 million shares every day.

Asked if humans are ever involved in the trading, Narang told correspondent Steve Kroft, "Humans are not involved in the trading because humans are way too slow to trade on the kinds of opportunities that we're trying to capture. We're trying to capture opportunities that exist for only fractions of a second."

The Tradeworx computers don't care where a stock is going to be trading next year, next month, next week or even tomorrow, because they are going to be in and out of it on the same day, in a matter of minutes.

"What's the point of buying and selling a stock that you hold for three minutes?" Kroft asked.

"Same objective that all other participants have in the market, is to make money. You buy low, sell high, that's how you make money," Narang said.

"And the computer will know when to buy and when to sell?" Kroft asked.

"Sure, the computer is monitoring real-time data and it knows what to do with that data and how to make decisions based on that," Narang replied.

What Narang and other high frequency traders tell their computers to do is to make a profit of a penny or less, 40 million times day.

They scan the different exchanges, trying to anticipate which direction individual stocks are likely to move in the next fraction of a second based on current market conditions and statistical analysis of past performance. But the computers have no real understanding of who these companies are and what they do.

The computer doesn't know or care whether a company is well managed. "It doesn't know who the CEO is or what that CEO's background is. Doesn't know the management team," Narang said.

"Whether he's going through a divorce?" Kroft asked. "Whether he's just been sued for sexual harassment?"

"Right. It knows information that you can quantify about the company," Narang explained.

Asked if it's all math, Narang said, "It's all probability and statistics - a procedure that you can define precisely."

The trading instructions are programmed into the computers with complicated mathematical formulas called algorithms. Narang showed us how it works with a simple, hypothetical example he uses for demonstration purposes.

"I'm gonna test a strategy where if a stock went down five percent for the past week, I'm going to buy $5 of that stock. And if a different stock went up ten percent last week, I'm going to sell $10 of that stock. And I'm gonna do that for every stock that's in my tradable universe simultaneously," he told Kroft.

"Which is how many?" Kroft asked.

"There's over 4,000 stocks, about 4,500 stocks," he replied.

The strategy, which could only be successfully executed with a high speed computer, would result in almost as many losing trades as winners, but over the past eight years would have produced a tidy profit - something that Narang and other high frequency traders have gotten used to.

Asked how successful he and his firm have been, Narang told Kroft, "We've had two or three days in a row where we lose money. But we've never had a week, so far, where we lost. We've never had a month that was a loser for us."

Just four years ago, high frequency traders accounted for 30 percent of the stock trades in the U.S. Today, estimates range as high as 70 percent. And institutional traders, like Joe Saluzzi of Themis Trading LLC, have come to believe that the game is rigged.

"How can you make money day after day? There was even one firm that said they made money four years in a row every single day. Well you have to be getting information that other people don't have, otherwise statistically that's an impossibility," Saluzzi said.

Actually, high frequency traders are getting the same market information that Saluzzi gets. They are just getting it a little bit sooner - it's only a few fractions of a second sooner, but if you are running supercomputers, Saluzzi says, it can be an eternity.

"What you're saying is the people with the fastest computers have an advantage? They get the best deals?" Kroft asked.

"Every time. Absolutely. There's no doubt about it. I mean, if they're spending that kind of money, and they're using that type of infrastructure, they're doing it for a reason. And it is to get a speed advantage, in that respect," Saluzzi replied.

It's not just the speed of the super computers that's important - it's also their physical location. The closer they are to the stock exchange's server the quicker they will be able to get critical market information.

Larry Leibowitz, the chief operating officer of the New York Stock Exchange, believes its massive new data center in Mahwah, N.J. will help the exchange regain some of the market share it has lost to electronic trading platforms. And he is busy persuading traders to lease space in the center's stark black boxes for their super computers.

It's called "co-location," a service that high frequency traders will pay tens of thousands of dollars a month for, and includes access to raw data from the exchange that is almost instantaneous.

"We're getting down to, you know, 'How fast can the electrons travel at this point?'" Leibowitz explained.

"They can predict the price of a stock before you can, because of the speed that they're using," Joe Saluzzi told Kroft.

"So, they actually see the trades before you do?" Kroft asked.

"They can see order flow coming into the exchanges before a regular person off of say a Bloomberg or somebody who doesn't have the co-location, the data feeds, and all the other sophisticated technology that they employ. Which is not cheap, by the way, it's extremely expensive to set these things up," Saluzzi said.

Asked how much faster they see it, Saluzzi said, "It could be a few milliseconds."

"How much of an advantage is a couple of milliseconds?" Kroft asked.

"Millions, if not billions of dollars a year," he replied.

That edge, Saluzzi claims, has made high frequency traders the new insiders on Wall Street, and he says he spots signs of predatory behavior every day. Saluzzi, who trades large blocks of stock for institutional investors, says the supercomputers are programmed to place and then cancel thousands of orders a second, trying to sniff out which way a market is moving in order to jump in ahead of big rallies and sell off before big declines. He calls them parasites who exploit a technological advantage to suck money out of the market and add no value.

Asked if high frequency trading raises capital for companies, Saluzzi said, "Absolutely not. If anything, it's distracting from the capital raising process."

"Do these high-frequency trades have anything to do with market fundamentals?" Kroft asked.

"Valuation is irrelevant. It's all about just moving the price up and down the ladder all day long. Each day is new. Each day starts fresh. So, you have to question the true valuation of the markets now," Saluzzi said.

Larry Leibowitz of the New York Stock Exchange says there is absolutely no evidence that small investors are being hurt by high frequency trading. Most of them, he says, don't care about pennies when they are buying and selling stocks. And they're in it for the longer haul.

"Look, there's always been charges for as long as trading has existed that people are front running orders, manipulating stocks. This is nothing new. I think now you add to it the element of the mysterious element of 'the computer' and it makes people even more mistrustful," he told Kroft.

Leibowitz and other proponents of high frequency, high speed computer trading say it has performed a valuable function: tripling volume, reducing stock spreads and transaction costs, and providing liquidity to the markets.

"Liquidity means that if you want to buy or sell a stock you could do it right away, and you could do it at a fair price. That's what liquidity means. And without short-term traders, there is no liquidity," Manoj Narang explained.

Traders like Narang say their presence in the market is making it cheaper and easier for everyone to buy and sell stocks, but regulators and lawmakers like Senator Ted Kaufman of Delaware have other concerns.

"Clearly, liquidity's way up. But what I say is, liquidity's always trumped by transparency and fairness. You can't have fairness if you don't have transparency," Sen. Kaufman explained.

Kaufman, who has both business and engineering degrees, says he is a big fan of technology but he thinks it's gotten way ahead of financial regulators' ability to monitor it. Right now, it's not even possible to determine for sure who is making high frequency trades or what they are telling their computers to do.

"We don't know what's going inside those boxes. There's all types of allegations about what's going on inside there. And basically what can happen is you can have these meltdowns where you can have a computer just go crazy and cause all kinds of problems," the senator said.

Which takes us back to the mini crash on May 6, and one of the scariest rides in stock market history when the Dow Industrials at one point plunged 600 points for no apparent reason.

Turns out it was triggered when a mutual fund's computer dumped $4.1 billion of securities on the market in a 20-minute period, which were then gobbled up by the computers of high frequency traders and sold almost immediately, sending other computers and traders heading for the exits.

"The events of May 6th scared people. I don't think there's any question about that," SEC Chairman Mary Schapiro told Kroft.

Schapiro had already proposed rule changes before May 6 that would allow regulators to track and tag high frequency trades and she is now considering further measures.

"Are you comfortable with computers making 50 to 70 percent of the trades on Wall Street?" Kroft asked.

"One of the concerns is, if one goes wrong, if it operates in an unexpected way, given market conditions, what's the impact of that algorithm that has behaved in an unexpected way, on lots of other investors in the marketplace?" Schapiro replied.

And Schapiro says it has happened since the May 6 crash, after circuit breakers were put in place that automatically halt trading in a stock that moves more than 10 percent in a five minute period.

"A number of times that those circuit breakers have been triggered has been because an algorithm operated in a way nobody intended for it to, causing a stock price to go wildly out of range," Schapiro said.

The crash contributed to the crisis in confidence on Wall Street. Since last spring, people have pulled $70 billion out of mutual funds and the biggest concern of Schapiro and Senator Kaufman is that average investors have lost faith in the integrity of the system.

"Is that correct?" Kroft asked the senator.

"Yes, that's true. Correct. And I'll give you an example. When I was at Wharton, [a] professor came and he said, 'You know, there's a river of wealth that runs through this country.' He said, 'A very small number of people know that it exists. Some people can stand on a high hill and see it off in the distance. Some people can get up on the edge. And there's other people are swimming in it.' That's the perception American people have about what's going on Wall Street right now. They believe there's a small number of people who are swimming in this river of wealth," Kaufman replied.

"There are a lot of people out there who think that the stock market is rigged. Rigged in the sense…that there are people out there who have advantages, the insiders, the big companies?" Kroft asked Larry Leibowitz.

"Right. Yep. And I think that we have to do a better job of, first, obviously making sure it's not the case," he replied. "But we can't be evasive about it. We have to make changes that make sense, that give people more confidence in the market, add more transparency, and make people feel like, 'This is a place I can trust my retirement savings to.'"

Produced by Tom Anderson