Investors in three of the biggest Dow Jones Industrial Average stocks were whipsawed by price swings that repeated every hour Thursday, fueling speculation the moves were a consequence of computerized trading.
Shares of International Business Machines Corp. (NYSE: IBM), McDonald’s Corp. and Coca-Cola Co. swung between successive lows and highs in intervals that began near the top and bottom of each hour, data compiled by Bloomberg show. While only IBM finished more than 1 percent higher, the intraday patterns weren’t accompanied by any breaking news in the three companies where $3.42 billion worth of shares changed hands.
Regulators have increased scrutiny of computerized strategies that rose to prominence in the U.S. after more than a decade of market structure reform. The Securities and Exchange Commission and Commodity Futures Trading Commission blamed a broker’s algorithm for setting into motion the events that caused the May 2010 market crash that briefly erased $862 billion from U.S. equities in less than 20 minutes.
“Somebody probably has software that’s running an algorithm that’s either selling in 30-minute intervals or buying,” Bruce W. Weber, dean of the Alfred Lerner College of Business and Economics at the University of Delaware, said in a telephone interview. “For the market value of Coke to be going up and down in this way, oscillating every hour, is a pretty disconcerting observation. This is not going to raise investors’ confidence in the mechanics of our market.”
John Nester, a spokesman for the SEC, declined to comment on the trading, as did Kent Landers, a spokesman for Coca-Cola.
Vineeta Durani of IBM didn’t return a call and e-mail requesting comment.
McDonald’s didn’t immediately make a representative available for comment.
“We’re not aware of any issues,” said Rich Adamonis, a spokesman for the New York Stock Exchange, where the companies are listed. “Nor were there any circuit breakers or other halts applied to those stocks. We didn’t receive any complaints in our market. Everything was fine.”
IBM fell 1.5 percent to $192.45 Friday, while McDonald’s slipped 1.3 percent to $91.58. Coca-Cola lost 0.7 percent to $77.03.
Apple Inc., which also had hourly price swings Thursday, dropped 1.6 percent to $604.30 Friday.
Algorithms are trading strategies that break larger buy or sell orders into smaller pieces over a specified period of time such as an hour or day. Traders at fund managers or brokers use them to mute price impact and mask activity as they try to scoop up visible and hidden orders spread across exchanges and alternative venues such as dark pools.
A volume-weighted average price, or VWAP, strategy seeks to buy or sell stock over a certain period, weighted for the number of shares traded at different levels. A time-weighted average price algorithm tries to buy or sell over set intervals of time. More aggressive and tailored algorithms exist to fine-tune trading based on market conditions such as volume, volatility or the price movements of indexes.
Together, the three Dow stocks represented 14 percent of the value of trading in the index yesterday. IBM, which reported results July 18, made up 8.1 percent, with $2.03 billion changing hands, more than every stock in the index except Pfizer Inc., the data show. Yesterday had the highest value traded for the three stocks since March 16, when $3.52 billion changed hands, and it is almost twice the 2012 average of $1.96 billion.
Apple traded $9.56 billion of shares yesterday, the most since July 11, data show. The stock rose about 0.5 percent six times at the start of the hour and lost part of the gains by the end, the data show.
“Every fundamentally oriented trader/investor is scratching their head,” Nicholas Colas, the New York-based chief market strategist at ConvergEx Group, said in an e-mail. “These are high-dollar stocks, and major weightings, relatively speaking, in indexes such as the Dow and S&P 500.”
About 10.4 million shares of IBM changed hands yesterday, twice the daily average of the past 12 months, data compiled by Bloomberg show. About 9.03 million shares of Coke changed hands, 25 percent more than the three-month average. The total number of shares traded in McDonald’s yesterday was 7.43 million, 13 percent more than the three-month average. The number of Apple shares traded jumped 72 percent yesterday from July 18. About 15.6 million shares of Apple changed hands or 7.3 percent below the average in the past three months.
“Volume was significantly higher in most of those names yesterday, which combined with the exaggerated price moves would indicate that there were large orders placed in the algo,” said Mark Turner, head of U.S. sales trading at New York-based Instinet Inc., which accounts for almost 5 percent of daily U.S. equities volume, said in an e-mail.
Coca-Cola’s intraday low was $76.50 at 9:58 a.m. The stock rose to $77.25 at 10:28, fell 0.7 percent to $76.71 at 11 a.m., rose 0.9 percent to $77.40 at 11:29 a.m., and fell 0.8 percent to $76.79 at 11:59 a.m. It hit successive highs at 12:29 p.m., 1:22 p.m. and 2:29 p.m., data compiled by Bloomberg show.
“It’s conceivable there was an algo involved but most don’t run on the half hour,” said Bernard Donefer, a professor at Baruch College who wrote the April 2010 article “Algos Gone Wild” for the Journal of Trading. “They run every couple of minutes, or seconds or even faster. That doesn’t make any sense. It doesn’t seem like something you would see with a normal algo.”
IBM, whose earnings topped analyst estimates, closed up 3.8 percent at $195.34, the biggest gain in the Dow. During the day, the stock climbed 0.77 percent or more to an intraday high six times, each time losing almost all of its gain in an hour. The shares reached highs of $196.59 at 12:28 p.m., $196.44 at 1:27 p.m. and $196.49 at 2:29 p.m., data compiled by Bloomberg show.
“It looks like some bad trading,” David Mechner, chief executive officer of Pragma Securities LLC, a New York-based firm that provides algorithmic trading tools and analytical services, said in a phone interview about the activities across the four companies. “There’s probably some kind of systematic trading in one interval that stops or reverses in the next. The unknown is what was driving that.”
A series of aggressive half-hour orders is a more likely explanation than a large buyer and seller alternately pushing the prices up and down, Mechner said. Gaming activity isn’t likely since that usually occurs at much shorter time frames and involves less financial risk, he said.
“Companies that make mistakes like that get punished by losing money,” Mechner said. “It’s not a threat in terms of systematic risk to the market. It was very odd to see swings that represent so much value that were so regular, but I think the market worked fine.”
The swings may have resulted from hedging long equity options since the contracts were due to expire today and the largest deals were at levels close to the stock price, increasing the risk their owners faced, according to Marko Kolanovic, global head of derivatives and quantitative strategy at JPMorgan & Chase Co. If the price of the stock underlying the option rises, the owner would need to sell shares to manage risk, or buy to manage the option position if shares decline.
The four stocks with repeating price patterns yesterday had the biggest net long options positions among S&P 500 Index companies, according to JPMorgan’s calculations, Kolanovic said in a note to clients today. The amount traded in the stocks was also consistent with what traders would have had to buy or sell, indicating that the patterns could be “almost entirely explained” by their need to hedge, he wrote.
“We believe that the price pattern of KO, IBM, MCD and AAPL yesterday was caused by hedging of options by a computer algorithm,” Kolanovic said in the note that referred to the companies by their ticker symbols. “It was likely an experiment in automatically hedging large option positions with a time- weighted algorithm that has gone wrong for the hedger.”
The SEC voted last week to require exchanges and the Financial Industry Regulatory Authority to build a single system to monitor and analyze trading activity on U.S. equity and options markets. The rule mandates a so-called consolidated audit trail to expedite surveillance across 13 equity exchanges, 10 options markets and more than 200 broker-dealers that execute stock trades away from public venues.
The effort is part of the agency’s response to the May 6, 2010, stock rout that temporarily sent the Dow down almost 1,000 points. The so-called flash crash was triggered by a mutual fund firm’s algorithmic trade that sparked the rapid selling of financial futures because it took into account volume but not price or time, according to a report released by the SEC and CFTC in October 2010.
“It’s probably disconcerting to people who want to feel stocks are always priced efficiently,” Mechner said about yesterday’s price moves. “But it’s well known that trading activity can push prices out of equilibrium and when aggressive trading stops, prices revert.”