What Can Be Done to Slow High-Frequency Trading?
Would it be possible to impose a speed limit on high-frequency trading? That is the question currently hovering in the air, after Mary Schapiro, chairman of the Securities and Exchange Commission, warned this week in New York that the SEC is planning new controls following the May 6 “flash crash”.
But as the debate intensifies about hyper-fast equity trades, investors and policymakers would do well to remember another point. As a fascinating paper from Andy Haldane, an official at the Bank of England points, what makes the flash crash interesting is that it was not an isolated incident: on the contrary, it epitomizes, in an extreme form, a bigger problem of speed in modern finance.
And while this “speed” issue has not garnered much attention in recent years – partly because most observers assumed that speed was good – it seems that a debate is long overdue. Not only does the financial system seem to have sped up dramatically in recent years, but this trend has caused destabilization in ways that go well beyond the “flash crash”.
The key issue at stake, Mr Haldane argues, lies not so much with computer models, but issues of human behavior. More specifically, he points out, neurological research suggests that the human brain has two contradictory instincts: part of it is hard-wired to chase instant gratification; however, another part of our brain also has the ability to be “patient”, and delay immediate gratification for future gains.
Now, one might have expected that during the course of evolution, humans would have moved from impatience to patience. However, Mr Haldane suggests this is not necessarily the case as far as finance is concerned.
On paper, most of the financial innovations in the past two centuries could – theoretically – have encouraged greater patience. The creation of liquid and deep markets, for example, has enabled pools of capital to be deployed to promote long-term investment.
Similarly, as corporate transparency has risen and information technology improved this has offered investors the ability to take well-informed, long-term decisions – if they choose.
But in practice innovation also has a darker, impatient side too: as markets have become deeper, and more liquid, that has enabled trading to become more frenetic; similarly, as information has become more frequently available, this has encouraged skittish, herd behavior.
Thus investors are increasingly demanding quicker returns. Equity churning has grown: whereas the average holding period for US equity holdings was around seven years in the 1970s, it is now nearer to seven months.
That appears to have promoted more market volatility: though equity prices were twice as volatile as fundamentals back in the 1960s, they have become between six and 10 times more volatile since 1990, with numerous miscorrelations.
And that in turn, has created a bitter irony, Mr Haldane argues: namely that while most of western society has long assumed that speed was tantamount to progress and efficiency, in truth these rising levels of speed, impatience – and short-termism – might have actually made the system less efficient, and rational than before.
Now, that conclusion will not come as a surprise to any investors who experienced the “flash crash” on May 6. Nor will it surprise anyone who has ever read a western lifestyle magazine; these typically rail against the way that life is “speeding up” in all manner of spheres.
But the question for an institution such as the Bank of England, or any other regulator, is whether anything can be done about this issue of speed; other than simply taking a luddite sledgehammer to computers?
Some vague ideas are now floating around. Lord Turner, head of the UK’s Financial Services Authority, is one of those who has floated the idea of introducing a “Tobin Tax”, or a tax on trading, to curb frenetic churn.
Mr Haldane, for his part, suggests policymakers might introduce incentives to promote long-term investments, such as giving more voting power to shareholders who retain equity stakes for a long period.
Ms Schapiro, for her part, raised another, more limited proposal this week: in a speech in New York, she indicated that the SEC is considering introducing a minimum “time in force” for orders, to stop high-frequency traders from frenetically canceling deals.
However, while Lord Turner and Mr Haldane’s ideas seem quite sensible, they stand little chance of flying anytime soon. Meanwhile, the Schapiro proposal barely scratches the surface of the problem.
To my mind, the real question which needs to be discussed – but which regulators are still ducking – is why ultra-fast trading is needed at all? What is actually gained by having deals struck at “one thousandth of a second”, as Ms Schapiro says? I would be interested to see some convincing answers.