sexta-feira, 28 de agosto de 2015
Gillian Tett: Welcome to a wild world of robot investing
This week’s turbulence in the markets was not just a reminder of the ever -growing importance of China’s economy; it was also testimony to how computers dominate the workings of the west’s stock exchanges.
Never mind that the Dow Jones index plunged by 1,000 points in just a few minutes on Monday morning (before later rallying). What was more startling was that the share price of stalwart American companies such as Apple, Home Depot or General Electric gyrated even more dramatically in minutes. * Meanwhile, the value of some exchange traded funds tumbled more than 30 per cent. So much for the idea that such funds are boring.
While it may take weeks before regulators understand why the plunges occurred, one reason for the swing is that automated computer programs have changed how markets function. The use of similar programs — such as high-frequency trading strategies — has expanded so rapidly that these are now estimated by the Securities and Exchange Commission to represent more than half of all US stock trades, and a big chunk of other asset markets.
Orders are being executed at lightning speeds in huge volumes. But there is another, often overlooked implication: these machines are being programmed to link numerous market segments together into trading strategies. So when computer programs cannot buy or sell assets in one segment of the market, they will rush into another, hunting for liquidity.
Since their algorithms are often similar (or created by computer scientists with the same training) this pattern tends to create a “herding” effect. If a circuit breaks in one market segment, it can ripple across the system faster than the human mind can process. This is a world prone to computer stampedes.
Some financiers insist this does not matter. Financial history amply shows that panic selling often occurs with humans too (the US stock market lost almost 90 per cent of its value in the three years after the 1929 crash). The good news about 21st-century computer stampedes is that while they are violent, they tend to be shortlived. By Tuesday the price gyrations seen on Monday had largely died down; similar recent wild bursts of volatility in bond markets also vanished fairly rapidly.
But the bad news is that when these computer stampedes do occur, small players and retail investors tend to suffer most. We do not yet know precisely who lost and gained most this week. But Douglas Cifu, head of Virtu, the world’s largest high-frequency trader, has revealed that Monday was one of the most lucrative days his firm has ever seen; other high-frequency traders have echoed this. Conversely, many of the investors who were trying to sell exchange traded funds at tumbling prices via their brokers on Monday were almost certainly retail players.
Little wonder that Jim Cramer, the CNBC television host who is popular with small investors, presented his show on Monday under the tag “Rage Against the Machine”; nor that Michael Lewis’s critical book about high-frequency traders was a best seller last year. For many western investors, this week’s events showed there is a yawning inequality in modern markets.
There are no easy solutions. In the past, banks smoothed trading flows by acting as market makers. But they have partly withdrawn from that role due to a regulatory squeeze. And nobody seems ready to kick computing trading programs out of the market, since in normal times they appear to provide the liquidity that banks no longer offer. Without high-frequency traders it would probably be more costly for investors of all sizes to trade.
In response, policymakers are now trying to make algorithmic trading more transparent and robust. After a market “flash crash” in 2010, the New York Stock Exchange introduced new circuit breakers, which temporarily stop trading when stocks gyrate too much. While these can sometimes calm markets, on Monday they may have created more panic: one reason the prices of exchange traded funds swung so bizarrely was that there were no prices available for the underlying stocks.
The result is that policymakers — and investors — are in a bind. Nobody wants to get rid of the robots — just this week BlackRock announced that it was purchasing a so-called “roboadviser”, to tap into swelling consumer demand for automated portfolio management. But nobody quite understands what the robots are doing to markets, let alone trusts them. What is crystal clear is that wild gyrations of the sort seen this week are now a central feature of our modern, robot-dominated markets. Human investors, stand warned.