Funds Flocking to Algos Signal Risks for Scottish Widows

Traditional fund managers are using computer programs to drive their trades as never before, creating the potential for large losses if the strategies aren’t implemented properly, according to Lloyds Banking Group Plc (LLOY)’s asset-management unit.

“There are an absolute plethora of algorithms out there,” Tony Whalley, who helps oversee 147 billion pounds ($223 billion) as head of equity dealing at Scottish Widows Investment Partnership in Edinburgh, said in a phone interview. “The more complex an algorithm, the more correct your assumptions about market conditions and what is going to happen over the period of that order have to be.”

More fund managers are employing software to try to obtain the best price for a trade by feeding buy or sell orders into the market gradually. European asset managers sent 43 percent of orders using algorithms last year, up from 25 percent in 2008, according to research by Tabb Group LLC, a consultancy firm based in Westborough, Massachusetts.

Investors using algorithms they don’t fully understand may be contributing to spikes in volatility and money managers must place controls on the bids and offers programs are able to place as a safeguard against errant trades, according to Whalley.

“There are occasions where you look at the market and there has been a sudden spike in the stock price,” said Whalley. “It is nearly always down to an algorithm that has been put in without a limit. To put an order into the market without a limit on it is plain stupid.”

HFT Differences

Scottish Widows and other traditional fund managers generally use algorithms in a different way to so-called high-frequency-trading firms, which employ computers to spot and exploit price discrepancies in a fraction of a second. HFT firms place their orders directly with exchanges while asset managers often will trade via a broker. Scottish Widows doesn’t develop its own trading programs, and tends to use software provided by its broker, according to Whalley.

“Speed is very low in terms of our list of priorities,” Whalley said. Still, the company has “no qualms” about dealing with HFT firms because they have different investment horizons, he said.

“They have a micro or nanosecond time scale in terms of how long they want to get their business done in,” he said. “We have a three-year view. If someone else is in there providing liquidity, we want to be able to access that.”

Flash Crash

HFT has come under increased scrutiny from regulators after the so-called flash crash of May 2010, when the Dow Jones Industrial Average (INDU) briefly lost almost 1,000 points. HFT algorithms buying and selling securities rapidly led to the sudden removal of liquidity from futures markets, kicking off a related plunge in stocks, a report by the U.S. Securities and Exchange Commission and the Commodity Futures Trading Commission said in September 2010.

A computer error at Knight Capital Group Inc. in August last year sent U.S. stocks swinging as much as 151 percent and landed the market marker with more than $450 million of losses that pushed the Jersey City, New Jersey-based firm to the edge of bankruptcy. Knight later agreed to be bought by Getco LLC, a closely held high-frequency trader based in Chicago.

Scottish Widows also trades on so-called dark pools, venues that match orders anonymously and don’t publish prices, when the fund manager wants to obscure its trading strategy from other market participants, Whalley said.

“They work well,” Whalley said. “They preserve our anonymity and they minimize the market impact that any particular order is going to have.”

It is more important than ever that fund managers employ dealers with the experience necessary to figure out the best route to market for orders, according to Whalley.

“Anyone who puts an order into the market without a limit on it deserves everything they get,” he said.

To contact the reporter on this story: Jonathan Morgan in Frankfurt at jmorgan157@bloomberg.net

To contact the editor responsible for this story: Andrew Rummer at arummer@bloomberg.net

Press spacebar to pause and continue. Press esc to stop.

Bloomberg reserves the right to remove comments but is under no obligation to do so, or to explain individual moderation decisions.

Please enable JavaScript to view the comments powered by Disqus.