image description

Automated Trading Leaving Retail Investors In The Dust

by Michael W. Stocker |
The greatest challenge for federal investigators sifting through the events of May 6 was the lightning pace at which the crash developed

While the Securities and Exchange Commission's and Commodity Futures Trading Commission's joint final report on the flash crash lays to rest some of the more outlandish theories for the market's roller-coaster plummet and recovery of last May, it also puts the spotlight on the perils of a new world of automated trading that is rapidly leaving retail investors in the dust.

The greatest challenge for federal investigators sifting through the events of May 6 was the lightning pace at which the crash developed. As the day dawned, investors were already skittish over worries about the European debt crisis-and by 1 p.m. in New York, negative market sentiment had driven down share prices for many stocks. The Dow Jones industrial average was down by more than 300 points by 2:42 p.m.

At that moment, something mysterious turned what was already a broad downturn into a full-blown panic. In the next five minutes, the market dropped another 600 points, representing the evaporation of a trillion dollars of market capital-the greatest intraday decline in the DJIA's 114-year history. Then, just as stunningly, the markets resurged 600 points just 10 minutes later.

Despite the rapid recovery, the flash crash fanned investor anxiety that current trading strategies are no match for market volatility. SEC Chairman Mary Schapiro has suggested that concern over the flash crash may have contributed to the recent withdrawal of retail investors from the stock market, pointing out that since the crash roughly $70 billion has been pulled out of domestic equity mutual funds. Only now, more than five months since the crash, have the broader markets returned to levels seen in the days prior to May 6.

Investors have good reason to be worried. During the 15 minutes of the crash all the familiar laws of equity trading seemed suspended. Share prices for eight major S&P 500 companies, including Accenture ( ACN - news - people ) and Exelon ( EXC - news - people ), fell to pennies, while shares in other companies, like Sotheby's ( BID - news - people ) and Apple ( AAPL - news - people ), shot to over $100,000.

The new report's conclusions about the causes for these gyrations will likely do little to allay investor concerns.

The SEC and CFTC have fingered a single trader's enormous automated sell order as the chief contributor to the crash. At 2:32 p.m. on May 6, a large mutual fund complex initiated the sale of some $4.1 billion in E-Mini Contracts, a type of stock market index futures contract traded on an electronic platform. Instead of manually entering the trade, however, the trader executed the sales through a computerized algorithm that sold the trader's position according to market volume, without regard to price or to time.

As the $4.1 billion position began to be liquidated, the sales triggered aggressive selling of futures contracts by high-frequency traders using computerized programs of their own, and this large volume of sales only increased the speed of the mutual funds' algorithmic trading. The problem quickly spread to the market for individual stocks as automated trading programs in the equities markets responded to declines they observed in the futures market. In effect, a feedback loop had developed between automated trading programs, at a speed that defied any human ability to intervene.

This rapid price free-fall highlights the complexity and perils of the fast-evolving securities markets. Computers are given free reign to trade at an inconceivably fast pace, and every fraction of a second saved on a trade can give investors an edge over competitors. However, the interaction between automated execution programs and algorithmic trading strategies can expose shortcomings in market liquidity in seconds and precipitate widespread crises.

The temporary strategy adopted by the SEC and the exchanges to address this problem is the use of "circuit breakers" to halt trading at times of extreme volatility. These new circuit breaker rules, in effect until Dec. 10, halt trading of any S&P 500 or Russell 1000 stock that rises or falls 10% or more within a five-minute period. It is hoped that pausing trading will provide time for humans to reinsert themselves into runaway automated trading by reassessing strategies and resetting algorithm parameters.

A long-term solution to the problem of automation-driven trading crises may involve changes in the way the markets are regulated as well. While the flash crash originated in the market for futures contracts, overseen by the CFTC, trading automation quickly drove the problem into the broader markets under the SEC's purview. To effectively prevent future flash crashes, the two agencies will have to work together closely to search for solutions that work across different markets, demanding a coordination of agency efforts that has sometimes been in short supply.

In the meantime, however, computer-driven stock price volatility is the new reality-and retail investors who lack the technological edge to keep up in this automated trading arms race may soon find themselves left behind.