7.2  High-Frequency Trading

High-frequency trading (HFT) is a strategy that uses algorithms and specialized trading systems to determine whether to buy or sell securities. The algorithms are mathematical formulas and computerized instruction sets that allow traders to execute a trade in a very short amount of time—we are talking nanoseconds or a billionth of a second. Computers that are close to exchanges or ECNs have an advantage over computers that are farther away because information takes time to travel. For this reason, high-frequency traders try to get as close as possible to the rest of the market, sometimes by having their computers “co-located” in an exchange. In that vein, high-frequency traders are always seeking the best “low latency.” Low latency is a measure of the total time it takes for the following:

• Information to reach a trader’s system

• For the algorithms to respond to the information

• The order to reach an exchange and be implemented

High-frequency traders usually make tiny margins on each order, sometimes just a fraction of a cent. So by having their lightning-fast systems close to the exchanges, HF traders can have access to information and be able to take advantage of arbitrage opportunities faster than non-HF traders. And because they may make millions of trades at a time, small margins can turn into big profits.

Proponents of HFT say that it provides more liquidity to the market, narrows bid-ask spreads, and lowers volatility. Researchers have found that the increase in HFT has lowered transaction costs for retail investors.

Critics of HFT say it exacerbates anomalies in the market. In fact, many have blamed HFT for the Flash Crash that occurred in 2010. HFT has also been criticized because many orders sent out by high-frequency traders are eventually cancelled. Critics of HFT also point out that the exchanges encourage HFT by paying HFT firms for order flow. This can be a significant source o