2nd Week Buttonwood-Not so fast

[이혜영]

Buttonwood

Not so fast

The risks posed by high-frequency trading

Aug 6th 2011 | from the print edition

• -->

FASTER is not always better. Who would want a doctor’s appointment that lasted 30 seconds or a whisky that had been matured for a week? But when it comes to the stockmarket, the acceleration of high-frequency trading (HFT) seems unstoppable. Trades on NASDAQ can be made in 250 microseconds, or one four-thousandth of a second.

Liquidity is assumed to be good for capital markets. If investors know it is easy to sell their holdings, they will be more willing to buy. This will lead to a lower cost of capital for business and thus more investment in the economy. HFT, which accounts for two-thirds to three-quarters of all Wall Street volume, seems to have led to smaller spreads (the gap between bid and offer prices).

In this section

• Why the tail wags the dog
• Reflections of reality
• The Corzine put
• ≫Not so fast
• The midget and the mighty
• High hopes, low returns
• Flat plan
• The revenge of Malthus

Reprints

But the stockmarket is also supposed to be concerned with the efficient allocation of capital. Companies with the best growth prospects should find it easiest to raise money. That needs a more considered analysis of corporate prospects than can be achieved in under a second. The average holding period for shares traded on the New York Stock Exchange has fallen to four months. Whereas Warren Buffett searches for companies with a sustainable franchise that trade below their intrinsic value, HFT might as well be buying and selling baked beans.

The size of spreads is not necessarily the only criterion of market efficiency. Traditional marketmakers earned a spread in return for the requirement to provide liquidity in all market conditions. The spread compensated them for the risk of being landed with unwanted inventory in periods of turbulence.

But as Andrew Haldane of the Bank of England pointed out in a speech* last month, HFT firms may be benefiting at the expense of other investors. This is an “adverse selection” problem. Instead of being better informed about a company, HFT outfits are simply seeing and acting on market prices sooner than competitors. “To be uninformed is to be slow,” he said. “These uninformed traders face a fundamental uncertainty. They may not be able to observe the market price at which their trades will be executed.”

Furthermore, as Bruno Biais of the University of Toulouse and Paul Woolley of the London School of Economics explain in a recent paper†, HFT firms indulge in “dark arts” like placing “spoof” orders to sell shares at attractive prices. These are designed to scare gullible traders into offloading their own shares, at which point the HFT investors withdraw their own spoof trades and snap up the real ones.

Profits of this sort can be viewed as “informational rents” earned at the cost of other market participants. But since these rents are made at the expense of the pension funds and insurance companies on which industry depends for long-term finance, it is not clear that the economy benefits from the process.

Worse still, higher trading volumes have also been associated with higher volatility. This is neatly illustrated by a paper‡ to be presented at the annual meeting of the American Accounting Association this month. The volatility of individual stocks can be the result of new information, such as a profit warning. However the authors look for examples of companies that have switched stock-exchange listings, joined the S&P 500 index or have dual classes of shares, where trading volumes differ but the fundamentals (profits, dividends etc) of the firm remain the same. They find that in all three cases trading volume is positively correlated with higher volatility.

Mr Haldane finds that both volatility and correlation between asset classes have increased in tandem with volume. The problem may be that, unlike marketmakers, HFT investors have no obligation to trade in difficult conditions. “HFT liquidity, evident in sharply lower peacetime bid-ask spreads, may be illusory,” he said. “In wartime, it disappears.”

The “flash crash” of May 2010, when the markets collapsed for no obvious reason, illustrated some of the dangers. As Mr Haldane points out, share prices ceased to have any information content (Accenture, a consultancy, briefly traded at one cent). Although the market righted itself quickly, regulators are debating ways to step in when prices go haywire. Doing nothing is like allowing Formula 1 drivers onto suburban streets.

* “The Race to Zero”, speech to the International Economic Association Congress, Beijing, July 2011
† “High Frequency Trading”, working paper, April 2011
‡ “The Dark Side of Trading”, by Ilia Dichev and Dexin Zhou of Emory University and Kelly Huang of Georgia State University, August 2011

Economist.com/blogs/buttonwood