High-Frequency Trading

How computer algorithms trade shares in millionths of a second, the arms race for speed, and the debate over whether it helps or harms markets.

A large share of trading in modern stock markets is carried out not by people but by computer programs. High-frequency trading, or HFT, uses powerful computers and algorithms to trade very rapidly, often holding positions for only seconds or less.

How it works

High-frequency traders use algorithms to:

Make markets: continuously posting bids and asks and profiting from the spread.
Arbitrage: exploiting tiny price differences for the same asset in different places, such as the same share on two exchanges.
React to news and data faster than humans can.

Because profits per trade are tiny, HFT firms trade in huge volumes.

The race for speed

Speed, measured as latency, is crucial. Firms pay to place their computers in the same data centres as exchanges’ systems, called co-location, to shave off millionths of a second. In one well-known example, a company built a new fibre-optic cable between Chicago and New York around 2010 to cut trading times by a few milliseconds. Firms later used microwave towers, since signals travel faster through air than through glass fibre.

Benefits

Research suggests HFT has helped narrow bid-ask spreads and lower trading costs for ordinary investors in normal conditions, as intense competition between high-speed market makers squeezes margins.

Concerns

Flash crashes: on 6 May 2010, U.S. stock markets fell sharply and recovered within minutes, an event known as the Flash Crash. Investigations found automated trading contributed.
Fairness: critics argue HFT gives an unfair advantage to firms that can afford the fastest technology.
Disappearing liquidity: high-speed market makers may withdraw during turbulence, just when liquidity is most needed.

The co-location controversy in India

The National Stock Exchange faced investigations after allegations that some brokers gained unfair early access to price data through its co-location facility in the early 2010s. SEBI later penalised the exchange and some officials. The case showed how milliseconds of advantage can matter greatly in modern markets, and why regulators pay close attention to fair access.

Regulation

Regulators have introduced measures such as circuit breakers, order-to-trade limits and rules requiring exchanges to test algorithms. Some economists have proposed frequent batch auctions, which match orders every fraction of a second rather than continuously, reducing the value of tiny speed advantages.

Thinking HFT affects only professionals

Because high-frequency traders are often on the other side of trades, their behaviour affects the prices and costs ordinary investors face, for better, through narrower spreads, and sometimes for worse, in sudden market swings.

Key takeaways

High-frequency trading uses algorithms to trade in huge volumes at very high speed.
Firms compete on latency through co-location, fibre cables and microwave links.
HFT has narrowed spreads in normal times but contributed to events like the 2010 Flash Crash.
Regulators and economists debate fairness and possible reforms like batch auctions.