THE GROWING dominance of computer-based trading strategies means that investors seeking to divine a simple fundamental explanation for daily stock-market movements are increasingly off-track, writes PROINSIAS O'MAHONY

More than 60 per cent of daily equity trading volume in the US now comes from so-called high-frequency traders – essentially computers that execute trades in hyperactive fashion, usually using strategies based on obscure and secretive mathematical formulas.

Five years ago, less than 25 per cent of trading volume originated in such fashion.

Executing thousands of trades in a day means that high- frequency trade funds think in terms of microseconds (onemillionth of a second) rather than seconds. In an effort to satisfy this need for lighting-speed executions, the New York Stock Exchange (NYSE) Euronext market recently announced system changes that will ensure trades are executed within five milliseconds. Just two years ago, the corresponding time was 350 milliseconds.

Even after this change, however, NYSE Euronext remains slower than rival electronic platforms that cater specifically towards algorithm-driven traders.

The move away from fundamentals-based trading can be seen by the job requirements listed by high-frequency trading firms seeking quantitative analysts.

“Candidates will have exceptional quantitative skills as well as programming skills and will write production-quality, high-reliability, highly tuned numerical code,” reads one recent advertisement.

Also required is “strong knowledge of computational numerical algorithms, linear algebra and statistical methods; and experience working with large data sets”.

Aimed squarely at mathematics and computer science graduates, such advertisements are conspicuous by their omission of any sort of reference to financial expertise.

This has shades of the furore within the banking crisis, where professionals with just those sorts of skills devised products that, it now emerges, few in the upper echelons of the financial services sector understood, undermining effective risk assessment.

There have been accusations that high-frequency trading has exacerbated market volatility.

“Computers are deciding when to buy and when to sell,” opined Michael Bloomberg last October.

“When the herd mentality gets automated, the stampede gets turbo-charged.”

This remains open to question. Few analysts, however, will disagree with the Wall Street Journal’s recent assertion that the increased automation means the stock market is “more prone than ever to large intraday moves with little or no fundamental catalyst”.

This raises questions as to the usefulness of much media reporting, with daily market summaries typically attributing market movements to one or two simple causal factors.

It is not always that simple. On June 22nd, the world’s media almost universally attributed heavy market falls to the fact that the World Bank had downgraded its global economic forecast for 2009. Rather than contracting by 1.7 per cent, as previously estimated, the new forecast was that the global economy would contract by 2.9 per cent.

However, the World Bank’s downgrade was first reported 10 days earlier, just before the meeting of G8 leaders in Washington. There may have been many reasons to sell on June 22nd, but the World Bank story was not one of them.

The desire for neatness and order can also be ascertained in explanations given for the frantic trading on February 27th, 2007.

That day saw the Dow Jones suffer its biggest one-day loss since 2001, with massive trading volumes overwhelming NYSE computer systems and causing a system blockage at one stage in the trading day.

Many explanations were offered, one of the most common being that former US Federal Reserve chairman Alan Greenspan had warned that the US “might” fall into recession. The reality was that Greenspan’s thoughts were widely reported the previous day and markets took barely a blind bit of notice.

Simplistic media explanations for ostensibly inexplicable market movements originated long before the rise of computer-based trading, however. One study looked at the 50 largest market moves in the US since the second World War and the accompanying media explanations. The explanations given were tenuous, to say the least.

Prof Robert Shiller, the behavioural finance expert and author of Irrational Exuberance, writes that they included “such relatively innocuous statements as ‘Eisenhower urges confidence in the economy’, ‘Further reaction to Truman victory over Dewey’, and ‘Replacement buying after earlier fall’”.

Quantitative trader and author of Practical Speculation Victor Niederhoffer is also scathing of the media’s formulaic approach: pick a theme, find a pertinent quote from a fund manager or economist, list a number of stocks that rose or fell as confirmation of this theme – “heedless of the worthlessness of anecdotal evidence” – and the journalist’s assertions “amount to nothing more than an invitation to trade the wrong way”.

Most of the time, market movement is little more than noise. Stop-loss orders may be triggered, technical levels breached, economic reports issued: traders hit “buy” and “sell” buttons for many reasons.

If media market summaries were of questionable veracity in the past, they are likely to become further compromised by the continued growth of high-frequency trading and the fact that fundamentals-focused fund managers increasingly will play second fiddle to computers.

Higher transaction costs mean high-frequency trading is a lesser force in European and Asian markets, although they too are ultimately likely to follow the US example. First-rate mathematicians like Jim Simons are likely to become increasingly prominent.

Under the media radar compared to the George Soros’s of this world, Simons is actually the most successful fund manager of the past 20 years: his $7 billion (€5 billion) Medallion fund has enjoyed annual returns in the region of 35 per cent since it began in 1988, even after charging a 5 per cent management fee and 36 per cent of profits.

With markets rising and falling as a result of complicated mathematical formulas, what is to become of the daily market summary?

One option is to follow the example of Haaretz, the Israeli newspaper that last month let poets and authors cover the day’s news as a one-off experiment.

“Everything’s okay,” the summary read. “Everything’s like usual. Yesterday trading ended. Everything’s okay. The economists went to their homes, the laundry is drying on the lines, dinners are waiting . . . Dow Jones traded steadily and closed with 8,761 points, Nasdaq added 0.9 per cent to a level of 1,860 points. The guy from the shakshuka [an Israeli egg and tomato dish] shop raised his prices again.”

A little bit of lyricism combined with an absence of dodgy and unsupported inferences – what’s not to like?