High Frequency Trading or HFT for short has been around pretty much ever since the financial industry found out what computers were and started incorporating them on exchange floors, big boiler room offices and various other parts of their business operations.
The idea is that modern finance is practically impossible without the fast-calculating, fast-processing and analyzing of various assets. If all of that was happening on the exchange floor 30-40 years ago, now it’s happening in an investor’s bedroom at probably
1000 times the speed.
However, that speed aspect is slowly but surely becoming a topic of concern and criticism in the larget sense of the financial world. The argument is that it’s simply way too fast and that people that have access to this trading strategy are
getting an unfair advantage over the average traders that try to make a buck on the markets.
Regardless of how you see it, HFT is a big part of Fintech, as it’s quintessentially dependant on fast computers and an open stock exchange. But before I start ranting on how HFT has become such a hot topic for debate, let’s first find out how it actually
works or what it is in general.
High Frequency Trading derives its name from the speed at which a specific computer or computers can process thousands of trades with large volumes. Something that would have taken a whole day to complete 30 years ago now takes just a fraction of a second
depending on where the server is located.
Let’s now dive into a more detailed explanation of how it works, and who are the people that make it work. And in order to do so, let’s tell an engaging story.
Meet Bob. Bob is a retail investor in the stock market. He’s an electrical really likes cars and knows quite a lot about them, therefore he’s decided that he wants to trade Tesla stocks. And by trading, I mean he wants to buy some, 1000 of them to be exact.
He quickly puts in the order on his stock exchange software and indicates the maximum amount he’s ready to pay per share. The
current price is $255 and $260 is the maximum that Bob is willing to pay (He really likes Elon Musk).
In order for this order to be processed, it needs to be a bit more manageable. The exchange server is programmed to break down this 1000 share order into two or ten pieces. Let’s go with the ten-piece version for now.
So, the exchange pretty much receives 10 orders of 100 Tesla shares at a maximum of $260, anything above it will not be credited to Bob. Once everything is in place, the first batch of 100 shares for $255 apiece is processed and credited to Bob with another
on its way.
But, in the process of processing this first batch, Dave’s HFT software manages to catch the transaction in the system.
Dave is the owner of an investment firm on Wall Street and a High Frequency Trader. He pays millions of dollars for real estate just to be as close to the exchange as possible, why? We’ll find out soon.
Once Dave’s HFT software notices the activity on the exchange, it tries to somehow identify the best case scenario with Bob. It immediately knows that Bob wants 900 more shares of Tesla, but doesn’t know what his maximum price is. However, through an algorithm,
the HFT software knows that there is some kind of markup to the current market price that all traders indicate in their maximum price, therefore it will start trying to guess it.
It goes for $270 but fails. It tries again with $265 but fails. A third time with $260 and it’s immediately processed. The HFT software has guessed the maximum price that Bob has indicated on his order.
After this, the HFT software quickly buys 900 Tesla shares for less than $260, and then immediately sells them to Bob for $260. Even if it’s a $1 profit, the overall trade will have earned Dave $900 in total without the fees. With fees, it would be around
$800 or so.
Even the $800 may seem small but consider this. All of what I’ve just talked about would happen in 0.3 seconds after Bob places his order. Pretty wild right? So much stuff happened in such a short time, and Dave managed to make $800 in a second when it’s
sometimes the monthly salary or even more of several countries.
Why is this morally grey?
There have been quite a lot of complaints about this method of trading, why? Because it’s pretty much available only for the rich.
Remember when I mentioned that Dave spends millions on renting real estate next to the exchange? He does this so that his HFT server is as close as possible to the stock exchange server. Meaning that no matter how strong the transmission is, his orders are
likely to be processed faster, even if it’s by just 0.000001 seconds. Those seconds are what get Dave to make that much in such a small amount of time.
It’s no secret that not everybody has millions to spare for real-estate as well as the human resource and equipment for HFT software, right? That’s why most exchanges are
considering to ban the activity completely.
Another reason is the unfairness of the market and price manipulation. If it weren’t for Dave’s HFT software, Bob would have bought more Tesla shares with his investment as very few orders would match his $260 maximum due to the current market price of $255.
Overall, this would affect the price of Tesla shares if it was a large-enough order made by an institution.
Should HFT be banned?
This is a topic that has been up for debate around a decade or so. Should people who have already achieved financial success waste the resources they’ve gathered on anything other than HFT’s lucrative nature? If Yes then why? Because of unfair competition?
It’s capitalism, the whole gig is about competition.
Although these arguments would make sense in a courtroom and have been making sense for the last few years, it’s still important to consider what kind of implications the continuation of such practices would mean for the financial industry. Would freelance
investing cease to be due to such unfair advantages? Or will it simply decrease in volume to weed out HFT software users?
Before all that happens, we may see multiple financial regulators simply place a ban on such activities in the near future.