High-Frequency Trading (HFT): How and Why It’s Taking Over The Markets

 

High frequency trading is getting a lot of publicity these days in the media. But exactly what is HFT and how does it work?

 

High frequency trading is a computerized trading platform which executes a large number of orders at lightning-fast speeds. Complex, pre-programmed algorithms analyze multiple markets and execute orders. The faster they do it, the better. (Typically, traders with the fastest execution speed will be more profitable).

 

Not all computer trading programs are HFT, though. And not all high frequency trading programs are predatory (more on how that works in a moment). But all HFT programs are designed to increase the liquidity in their traded markets — this can be both good and bad.

 

High-frequency trading became popular when exchanges began offering incentives for companies to add liquidity to the market. “Adding liquidity” means having orders in the order book ready to accept execution. For example, the New York Stock Exchange has supplemental liquidity providers (SLPs) who add liquidity in an assigned security and are awarded a financial rebate accordingly. That SLP financial rebate is proposed to be raised from $0.0027 to $0.0032 per share as of April, 2012.

 

It may not seem like a lot, but multiply even $0.0027 by millions of shares per day and that’s a very lucrative incentive for high frequency trading.

 

There are other ways for high frequency traders to make money too. Here’s a predatory example:

 

Let’s say a mutual fund wants to buy 100,000 shares of XYZ trading at $10.00/share. The fund doesn’t want to pay more than $10.10/share and therefore places a limit order at $10.10.

 

So far, so good. But some exchanges allow high frequency trading firms to see new market orders milliseconds before anyone else (for a fee). Such a firm will see a flood of demand for XYZ stock and quickly estimate the limit price after buying some XYZ themselves.

 

The high-frequency trading algorithm can buy 100,000 shares of XYZ at $10.00 and start issuing small sell orders at prices such as $10.50. An order at $10.50 would probably get rejected because it’s higher than the limit price set by mutual fund. But after that it would try again at $10.25 (also rejected) and finally another small order for $10.05 which would get filled. A well-written algorithm would continue to send small orders until the limit price is determined to be $10.10 … and then flood the market with a large order at that price.

 

The profits made by this particular predatory algorithm are made at the expense of the mutual fund, which has paid a higher price than it really needed to pay. All of this could happen in less than a second, by the way.

 

And even though the profits per share are measured in pennies, the high volume of trading means huge profits (remember that there could be hundreds of orders from the various mutual funds or hedge funds each day).

 

Even high frequency trading algorithms which aren’t predatory can still make nice profits from arbitrage and the financial rebates offered by exchanges. This is probably why HFT firms represent only 2% of the trading firms in the US, but 70% of the trading volume according to the TABB Group. It really is that profitable if you’ve got the computing power and the algorithms to make it work.

 

It’s not just stocks, either: high frequency trading occurs in equities, commodities, ETF’s and almost anything traded electronically. So is this a good thing or a bad thing?

 

High frequency trading is just a tool, in the end. And as with any other tool in the world, HFT can be used positively (arbitrage, ensuring a liquid market) and negatively (predatory practises that cost other firms profits without their knowledge). Which usage will dominate the markets depends on how regulators decide to approach the issues.

 

However, it’s a safe bet that the “good” aspects of high frequency trading will always be permitted (and even encouraged) to ensure easy fills on orders and to correct mispricing between markets. So if you’re interested in trying HFT for yourself, try to create algorithms that work on those principles. You’ll still be able to make money with a clever program, and yet feel good about it at the same time.

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!