What is high-frequency trading (HFT)?

High-frequency trading is kind of like a high-speed train...

A typical train ride may take a while, but a high-speed train can take you to your destination much faster. Similarly, high-frequency trading can improve the market’s efficiency, connecting buyers and sellers at more advantageous prices.

High-frequency traders may profit off two primary factors—1) their trading volume and 2) their speed.

Here’s what happens: Complicated computer algorithms automatically hunt for trends across many different markets, aiming to notice movements before anyone else. They’re not looking for trends other investors might care about (e.g., companies that may be overvalued, sectors that might take off, or an impending recession). Instead, they’re hunting for thousands of tiny opportunities, like miniscule price differences for the same asset in different markets.

Since the differences are typically small, trading may lead to a small profit. Fluctuations can happen in the blink of an eye. But what they lack in size, they make up for in volume. Of course, the risk of losses is also significant.

For high-frequency trading firms, the biggest perk is obvious: It’s profit. Algorithms can keep an eye on a huge number of securities, markets, and exchanges and make thousands of trades per second. This may help a trader earn a profit based on small price movements.

Some supporters of high-frequency trading also assert that the practice improves market liquidity—that is, it makes it easier to find a buyer or seller for a security at a given price. High-frequency trading may support this by drastically increasing the number of trades happening and how fast they occur. Greater liquidity means investors face less of a risk that there won’t be a willing buyer or seller on the other side of a transaction. However, some scholars have argued that high-frequency trading may reduce liquidity.

Some argue that high-frequency trading firms are not competing with individual investors, but rather benefit them through a symbiotic relationship. By offering securities at lower prices and subsidizing trades at retail brokerages, high-frequency trading firms can lower costs for individual investors.

High-frequency trading is sometimes controversial. Some critics argue that the practice benefits large financial institutions at the expense of individual investors or smaller firms.

For high-frequency traders, the risk of losses can also be significant. Compared to long-term investing, chasing short-term market movements involves an even greater chance of losing money. With so many trades happening so quickly, that risk is multiplied.

Some also blame high-frequency trading for causing price swings or even making the markets more likely to crash. For example, high-frequency trading helped trigger the “Flash Crash” of May 6, 2010, when major US stock market indexes plummeted and then partially recovered within an hour. (When all was said and done, nearly $1 trillion in market value had been wiped away.)

Some high-frequency traders have gotten in trouble for illegal practices. For example, layering is when a trader uses an algorithm to place multiple orders for a security at different prices to make it seem like there’s a lot of interest in buying or selling it. The trader then places new orders to benefit from the newly inflated or deflated prices and cancels their original orders. Nonetheless, high-frequency trading is a legal and widespread practice.

Before the latter part of the 20th century, securities traded in person — Buyers and sellers physically showed up on the floors of stock exchanges and used shouting and hand signals to close transactions. Starting in the mid-1970s, computerized trading allowed traders to buy and sell securities electronically. People no longer had to appear on the trading floor, and trades could be executed much faster. By the 1980s, virtually all stock trading took place electronically.

Starting in the late 1990s, advances in technology led to the emergence of algorithmic trading. This involved programming computers with pre-set instructions to execute trades based on certain variables, like time and price. This type of trading took advantage of the fact that computers could make these kinds of trades much faster than humans could.

Over the next 20 years, the rise of high-frequency trading has been fueled by ever-faster computing speeds and advances in artificial intelligence. Firms emerged that focus exclusively on this strategy, and high-frequency trading now makes up around half of trading volume in the U.S. stock market.

High-frequency trading algorithms do much of what humans used to do — just faster. For example, human traders have long used strategies like arbitrage (buying and selling assets like securities or currency to exploit price differences) or market making (trying to profit from differences in the bid-ask spread of a certain security, thereby adding liquidity to markets). High-frequency trading algorithms can carry out these strategies extremely quickly, which some say makes markets more efficient and stable.

Most brokerage firms serving retail investors today receive payment for order flow (PFOF). Here’s how it works: When a trader places an order, the firm sends it to a high-frequency trading firm (aka market maker) that completes the trade. In exchange, the firm gives the broker a PFOF (how much is based on the terms they’ve agreed on). These firms usually offer better prices for securities than you can find on the stock exchange.

Although PFOF has attracted some scrutiny in recent years, the practice can be a big boon for retail investors: They may get a better price for securities, benefit from more liquidity in the market, and have the opportunity to trade with low or zero commissions (since market makers are essentially subsidizing the cost).