In our web poll last week we asked about high-frequency traders, in response to comments made by Andy Haldane, the Bank of England’s head of financial stability, who compared last May’s flash crash to a physical catastrophe that we ignore at our peril.
In a speech he delivered in Beijing earlier this month, Haldane said that the drive to achieve zero latency, where trading converges at the speed of light, has destabilised markets and needs to be reined in. “There is nothing normal about recent deviations in financial markets,” he told the International Economic Association’s 16th World Congress. “The race to zero may have contributed to those abnormalities, adding liquidity during a monsoon and absorbing it during a drought.”
Several studies have found just the opposite — that new liquidity providers such as high-frequency traders bring execution certainty and price efficiency, which helps to narrow bid-ask spreads. But these studies are flawed, according to Haldane. By smoothing out volatility to get rid of abnormal deviations, they distort the very thing they are supposed to be studying.
“Normalising volatility might normalise abnormality,” he said. “It risks falling foul of what sociologists call ‘normalisation of deviance’ — that is, ignoring small changes which might later culminate in an extreme event.”
That could explain what happened on May 6 last year, during the flash crash, when hedge funds’ micro-liquidity disappeared down fibre-optic cables at close to the speed of light.
“Far from solving the liquidity problem in situations of stress, HFT [high-frequency trading] firms appear to have added to it,” he said. “And far from mitigating market stress, HFT appears to have amplified it. HFT liquidity, evident in sharply lower peacetime bid-ask spreads, may be illusory. In wartime, it disappears.”
Some academics have theorised that markets display fractal geometry. Cramming more trading strategies into smaller slices of time, they say, also crams in more abnormalities — because markets are “self-similar” on this scale. More trading means more problems.
During the flash crash, for example, algorithms executed 27,000 transactions of the S&P 500 E-mini futures contract in 14 seconds. Yet when those trades were netted off, just 200 contracts were bought.
Such madness also drives away the regulating effect of ordinary investors, who tend to halt trading when the robots take over and prices go haywire. That isn’t very significant during a micro-blip that lasts a few seconds, but it could mean that such blips have the capacity to run out of control. And, according to Haldane, as markets become more interconnected across assets, the risk of contagion rises.
In short, this is a call for more regulation. It is even a persuasive one, but our readers aren’t buying it. Slightly more than half of respondents said that high-frequency trading has a positive effect on markets, so perhaps the robots will be spared from the crusher for now.