How the Machines Took Over Wall Street

Have human beings lost control?

On May 6th, 2010, in the space of one-half hour, the Dow Jones Industrial Average (DJIA) dropped almost 1000 points. Two DJIA component stocks, Procter & Gamble and 3M, lost over 30% of their value in 15 minutes, while shares of Accenture went from $40 to a penny before recovering. What happened? Several theories were promulgated, from the "fat finger" trader error to "multiple erroneous trades." The common denominator? Advanced technology: computerized, high-frequency trading (HFT) that allows for the buying and selling of millions of shares in a matter of milliseconds. Completely automated trades made by super-computers, absent human interaction.

Computerized trading has been around for forty years. In 1971, NASDAQ became the first electronic stock market, allowing dealers to offer price quotes for securities. In 1976, the New York Stock Exchange (NYSE) initiated the use of its "designated order turnaround" system (DOT), which facilitated the routing of buy and sell orders previously handled by buyers and sellers standing next to each other on the exchange floor. In the 1980s, markets become fully electronic, and a new strategy called "program trading" was instituted. Program trading is defined as "any trade involving fifteen or more stocks with an aggregate value in excess of $1 million," and many people are convinced that it was the principal culprit or a substantial contributor to the October 1987 crash, where the market lost 22.6 percent of its value in one day.

In the '90s, electronic communications networks (ECNs) emerged, allowing traders to conduct business outside the boundaries of traditional exchanges and traditional business hours. This was made possible when the Securities and Exchange Commission (SEC) approved a regulatory framework for "alternative trading systems" due to the "important role of technology, and the increasing competition, in today's securities markets." ECNs spawned the growth of computer systems which use algorithms to facilitate trading.

In 2001, algorithm trading was further buttressed when markets began quoting stock prices in decimals instead of fractions. Thus the minimum size of a stock price went from 6.25 cents (one sixteenth of a dollar) to a penny, leading to increased liquidity. Finally in 2005, the SEC enacted a series of rules called the "Regulation National Market System," which facilitated the emergence of high frequency trading. High frequency trading may now account for as much as 60-70 percent of the volume on major U.S. stock exchanges. Some speculate it may be even higher than that.

“This is where all the money is getting made,” said former chairman and chief executive of the New York Stock Exchange William H. Donaldson back in 2009. “If an individual investor doesn’t have the means to keep up, they’re at a huge disadvantage.” Joseph M. Mecane of NYSE Euronext offered a vivid assessment of reality. “It’s become a technological arms race, and what separates winners and losers is how fast they can move,” he said.

Speed is indeed the name of the game. A two-tenths of a second advantage can be the difference between success and failure, allowing traders to buy or sell millions of shares, milliseconds ahead of the pack. The profit or loss on each trade may amount to tenths of pennies, but when those trades number in the millions over the course of a single trading day, tens of million of dollars can be won or lost.

Speed is so critical that a 40,000 sq.ft. facility in Mahwah, New Jersey -- a location chosen due to its proximity to Wall Street, but away from nuclear power plants, geological fault lines and flight paths -- was constructed so trading firms could "co-locate" their super-computers next to those of the trading exchanges in order to gain microseconds worth of speed over their competitors. "We're getting down to, you know, 'How fast can the electrons travel at this point?'" said Larry Leibowitz, chief operating officer of the New York Stock Exchange, who has spent time promoting Mahwah to NYSE clients. And then there's the advancement of computer technology itself. Vendors such as Hardcore Computer and Supermicro® UK, sell "submersion cooling technology," which enables them to "over-clock CPUs," allowing those processors to run faster, cooler and longer.

HFT has its proponents and detractors. Proponents believe HFT adds liquidity to the market and lowers the spread between the bid and ask prices. “High frequency trading has introduced a massive amount of competition, and competition is the only mechanism that can and will keep it honest," contends Eric Falkenstein, former portfolio manager and author of Finding Alpha (2009). "All it takes is ten smart people competing with each other to whittle profits almost down to nothing, all the while increasing the depth and lowering the spread to retail investors.”

HFT detractors believe the strategy has resulted in blatant market manipulation. One of those manipulative tactics is an abuse known as "quote stuffing." Massive amounts of orders are entered into an exchange, but then quickly withdrawn before any trades can be executed, a practice which is ostensibly illegal. But because the super-computers of competitors are still forced to process the data, they are slowed down, losing their competitive edge as result. Columnist Tyler Durden of explains:

If you could generate a large number of quotes that your competitors have to process, but you can ignore since you generated them, you gain valuable processing time. This is an extremely disturbing development, because as more HFT systems start doing this, it is only a matter of time before quote-stuffing shuts down the entire market from congestion.

These stuffed quotes can also have "shapes" that effectively manipulate the National Best Bid and Offer (NBBO) requirement -- the SEC rule that brokers must guarantee customers the best available ask price when they buy securities, and the best available bid price when they sell securities. As a result, detractors contend, much of the so-called liquidity provided by HFTs may be totally illusory.

Market regulators are apparently more concerned than ever. Last Friday, it was revealed that the Financial Industry Regulatory Authority (FINRA) and the Securities and Exchange Commission want high-frequency trading firms to turn over details regarding their trading strategies -- and, in some cases, their highly secret and often very expensive algorithms as well. "It's not a fishing expedition or educational exercise. It's because there's something that's troubling us in the marketplace," said Tom Gira, executive vice president of FINRA's market regulation unit.

The SEC has ostensibly adopted some new rules to prevent a repeat of the May 2010 flash crash mentioned above. Large traders, as in a person or company whose transactions in exchange-listed securities exceed two million shares or $20 million on any day, or 20 million shares or $200 million per month, will be assigned unique ID numbers. The traders will then supply those numbers to their broker-dealers, allowing the SEC access to those records if something goes amiss. The SEC is also proposing similar measures for dark pools, which are alternative trading venues that find bid/ask matches for blocks of shares without revealing the orders or the identities of the institutions involved. "Circuit breakers," which would temporarily halt trading in stocks that fall or rise 10% or more in five minutes, were also part of the mix. But those rules only apply to the S&P 500 and the Russell 1000 exchanges currently.

Critics consider these measures largely toothless and outdated, and unable to prevent another flash crash from occurring. Both articles referenced here accuse the SEC of institutional paralysis and calculated ineffectiveness, respectively.

Perhaps the SEC might want to examine suggestions put forth here and here. The former would be the imposition of a "minimum active quote latency," meaning no one would have a speed advantage over anyone else. This would eliminate the rationale for quote stuffing, which in turn, would essentially eliminate HFT. This would undoubtedly incur the wrath of many traders. The latter is a less onerous proposal, aimed at curbing quote stuffing by imposing a tax on order cancellations, thus discouraging the practice.

Whatever FINRA or the SEC does, one thing is certain: unless some sense of trust is restored into the market, investors will flee. After the 2010 flash crash, investors pulled $13.4 billion out of the market. And such skittishness is not limited to small investors. Consulting firm Grant Thornton reveals a 15-year decline in the number of initial public offerings by small companies.

A broker who wishes to remain anonymous explained that the kind of volatility we have seen in the market recently, such as the unnerving intra-day swings of several hundred points in both directions, may be the "new normal" within two to three years. Stock market volatility is measured by the "Volatility Index" (VIX), which hit a 52-week high following S&P's downgrade of America's credit rating. Such ongoing volatility, the broker contends, would drive most small investors, whose portfolios represent their retirement nest eggs and other critical savings, out of the market completely.

For the record, the official explanation regarding the flash crash on May 6th, 2010 is that the cause was a single firm's attempt to sell $4.1 billion in specialized futures contracts. This set off a "cascade of selling" by HFT super-computers, in turn causing one of the most volatile trading days in market history.

Joe Saluzzi of Themis Trading LLC, who characterizes HFT traders as the "new insiders" on Wall Street, gets to the essence of the problem. "Valuation [of a company] is irrelevant," he warns. "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." That's a troubling statement, which begets an equally troubling question: what does the future hold for stock exchanges whose principal reason for existence, aka the capitalization of companies, has become largely relevant?

The daunting new reality? Technology has likely turned trading activity into little more than a sophisticated computer game.