Securities and derivatives trading in the financial markets has been undergoing significant changes since the middle of the past decade. More and more frequently, computer programs are superseding human judgement in deciding what securities to buy or sell, when to buy or sell them and at what price. In a matter of milliseconds, special algorithms can analyse huge chunks of data and generate hundreds of orders. These traders are using direct and high-speed electronic access points to exchanges. Known as high-frequency trading (HFT), this now makes up nearly 50% of the trading volume in the most liquid segments of the US and European markets.

In recent months, a spate of severe price turbulence which cannot explained by the fundamentals has pushed HFT to the forefront of public and regulatory attention. Observers have blamed these market events on HFT activity and are debating the impact of these market players on the ability of financial markets to cushion shocks. In the current issue of its Monthly Report, the Bundesbank therefore investigates what effect the increasing speed of financial market activities is having on the capital market. Its research focuses on the impact of HFT activity on price efficiency, price volatility and liquidity provision in various market phases. On that basis, the Bundesbank’s economists studied copious data from DAX and Bund future contracts at the microsecond level.

HFT can amplify volatility in the short term

The Bundesbank’s results show that active, ie liquidity-absorbing market players take up a greater share of trade in periods of heightened volatility. At the same time, passive HFT market players, ie those that provide liquidity, typically keep a low profile by deleting trading orders, thereby reducing the supply of liquidity. According to the Monthly Report, one outcome of these different types of behaviour is that, in phases of market turbulence, the risk of excessive volatility increases, thereby provoking market turmoil.

Acting quickly improves price discovery

The Bundesbank has also examined the behaviour of high-frequency traders upon receiving important news, using US labour market data as an example. It finds that, especially when the markets are calm, high-frequency traders increase efficiency - as HFT causes news to be priced into market prices very quickly. However, according to the Bundesbank’s economists, this improvement is in the microsecond range - its economic value is therefore difficult to measure.

Study offers points of approach to regulation

The Bundesbank’s results provide possible starting points for the regulatory debate on HFT. They reveal the importance of setting incentives for passive high-frequency traders to continue providing liquidity even during periods of heightened stress. In addition, the excessive short-term volatility, to which active high-frequency traders contribute, could permanently discourage slower market players from providing sufficient liquidity during such phases. Measures to delay all market players’ ability to respond by fractions of seconds are already under discussion. According to the report, "This could, to some extent, offset the competitive disadvantages of slower market participants without perceptibly impairing technical progress on the trading platforms."