Content area
Full Text
INTRODUCTION
The trading of financial assets consists of taking a position on a particular asset and reversing it subsequently at a different price such that profit is generated. Predominantly, this exercise consists of taking a long position on an asset (the asset is bought) at a certain price, then the position is unwound (the asset is sold) at a higher price. 1 The difference between the selling price and the buying price is the profit generated from the operation after allowing for transaction costs. The time elapsing between taking a position and unwinding it (the holding period) could be very long, long, short or very short. When the holding period is very short (hence trading is frequent) and the transactions are initiated and executed by a high-speed computer running complex algorithms, the operation would constitute high-frequency trading (HFT), which is also known as 'high-speed trading'.
The holding period associated with HFT can be a fraction of a second, hence thousands or tens of thousands of trades are executed each day, but it can be as long (or as short) as a few hours. Hence, HFT is basically (an extreme version of) intra-day trading, which means that high-frequency traders do not hold overnight positions. The objective of HFT is to capture small profit per unit of the asset traded - by doing that thousands of times on big positions, significant profit can accumulate. As the concept is used currently, the necessary condition for HFT is frequent trading while the sufficient condition is the use of a high-speed computer to execute the trades, as the frequency of trading in a modern sense cannot be handled manually or by using a 1980s computer.
The basic concept of HFT has been known since the 1980s when several stock exchanges first decided to experiment with electronic trading. As a matter of fact, the crash of October 1987 is attributed, at least partially, to HFT.2 Since then it has become a commonly-used term, as the practice grew in terms of scope, speed and complexity, given the growing power of computing and progress in the design of complex trading rules by hard core mathematicians and statisticians. The International Organization of Securities Commissions (IOSCO) makes this point by suggesting that 'algorithms...