Jan. 8, 2015

New keyword

High Frequency Trading (HFT)

Trading called High Frequency Trading (HFT) which repeats buy and sell orders at a high frequency in milliseconds, has become mainstream in recent trading markets.

Takahiro Sasaki, Advanced Financial Engineering Center, NTT DATA Financial Solutions

What is HFT?

High frequency trading performed at high speed by tools that employ algorithms is called High Frequency Trading (HFT). However, there is no strict definition of HFT at present. In the report of the IOSCO (International Organization of Securities Commissions), it indicates that “defining HFT is difficult and there is no single agreed definition”Ref. 1. This report mentions the features presented by the trading participants of HFT, and the following features are listed as an example.

  • It is latency sensitive. The implementation and execution of successful HFT strategies depend crucially on the ability to be faster than competitors and to take advantage of services such as direct electronic access and co-location.
  • It is characterized by a high daily portfolio turnover and order to trade ratio (i.e. a large number of orders are cancelled in comparison to trades executed)
  • It usually involves flat or near flat positions at the end of the trading day, meaning that little or no risk is carried overnight, with obvious savings on the cost of capital associated with margined positions. Positions are often held for as little as seconds or even fractions of a second;
  • It involves the use of sophisticated technological tools for pursuing a number of different strategies, ranging from market making to arbitrage;
  • It is a highly quantitative tool that employs algorithms along the whole investment chain: analysis of market data, deployment of appropriate trading strategies, minimisation of trading costs and execution of trades;
  • It is mostly employed by proprietary trading firms or desks

In recent years, HFT trading has become mainstream in the major exchanges around the world, and the Advanced Financial Engineering Center of NTT DATA Financial Solutions has also selected HFT as one of the subjects of research.

Realized High Speed Trading Environment

The need for high speed and high frequency by the trading participants accelerated the improvements in the performance of the systems in exchanges, which further accelerated improvements in the speed of the trading participants. Tokyo Stock Exchange also released the new trading system Arrowhead in 2010. Although the average response time for an order was taking 2 to 3 seconds before Arrowhead started, it has now been reduced to an average of 2 milliseconds after its release. Subsequently, the performance of Arrowhead has continued to improve, and in 2012, the response performance achieved an average of 1 millisecond. Arrowhead will be renewed again in 2015, and is expected to achieve an average of 0.5 millisecondsRef. 2.

When the 1 millisecond demanded by HFT is converted into a distance both ways at the speed of light, it is about 150 km one way. This is equivalent to the distance from Tokyo to Shizuoka-shi. When the speed reaches this level, the geographical distance to the exchange must also be taken into consideration, not only the performance of the server and network systems. For example, if a server is installed in Osaka, the achievement of a 1 millisecond response time is physically impossible, because one way to Tokyo is about 400 km.


Figure1. Within 150 km Circle from Tokyo
It takes about 2.7 milliseconds to travel 400 km both ways between Tokyo and Osaka, even at the speed of light

In response to the need to minimize the geographical latency, most of the exchanges are providing “co-location” services where servers of trading participants are installed in a data center of an exchange (Figure 2)Ref. 3.


Figure2. Co-location
Service which allows installation of algorithmic trading AP server in systems of exchanges

Arguments on Effects of HFT

Although various studies have been conducted on what effect the increase of HFT has on the trading market, a unified opinion has not yet been determined. While there are opinions where the increase of HFT is providing liquidity in the market, there are also opinions on the possible deterioration in the quality of liquidity which will make the markets unstable, like the flash crash that occurred in the U.S. in 2010.

Since system failures caused by large volume orders and cancellations by HFT have occurred, the arguments to regulate HFT have been increasing. In the EU, mandatory testing and approval of algorithms by audit authorities were incorporate on the trading participants of HFT in the revision of MiFID II in May 2014 (Markets in Financial Instruments Directive II)Ref. 4. Close attention must be paid to the movements of each country in the future.


Authors Profile

Takahiro Sasaki, Advanced Financial Engineering Center, NTT DATA Financial Solutions
He was engaged in the engineering development concerning software production technology in NTT DATA Corporation, and was subsequently transferred to the Advanced Financial Engineering Center of NTT DATA Financial Solutions Corporation. Currently, he is responsible for the analysis of financial data, and the calculation model design of derivatives.

New keywords

  • High Frequency Trading (HFT)
  • Algorithm trading