Algorithmic trading and high-frequency trading
Algorithmic trading, whereby orders are entered, modified and cancelled by computer, carries various risks. For example, a high number of order entries, modifications or cancellations within a very short space of time could overload trading systems. Algorithms may also react to market events and trigger additional algorithms as a result, which may in turn trigger even more algorithms (cascade effect), leading to an increase in volatility.
On this page:
- High-frequency trading
- Authorisation requirement for high-frequency traders
- System and risk controls
- Identification requirement
- Appropriate order-to-trade ratio
- Other provisions
In order to counter these risks, the legislature enacted the Act on the Prevention of Risks and Abuse in High-frequency Trading, or High-frequency Trading Act (Hochfrequenzhandelsgesetz). The Act entered into force on 15 May 2013 and contains provisions relating to high-frequency and algorithmic trading which apply to both trading participants and trading venues. It is based on the new European provisions on algorithmic and high-frequency trading, which are envisaged as part of the overhaul of the Markets in Financial Instruments Directive (MiFID).
High-frequency trading is a form of algorithmic trading and often involves a large number of order entries, modifications or cancellations within microseconds. High-frequency traders seek to be as near as possible to a trading venue's server in order to derive speed advantages from the short distance the signals need to travel.
The following describes the main provisions of the High-frequency Trading Act (Hochfrequenzhandelsgesetz):
Authorisation requirement for high-frequency traders
High-frequency trading is subject to an authorisation requirement according to the German Banking Act (Kreditwesengesetz – KWG). The authorisation requirement applies to all direct and indirect trading participants in organised markets or in multilateral trading facilities (MTF) in Germany who trade using high-frequency algorithmic trading techniques, where no service to third parties is rendered. High-frequency algorithmic trading techniques are characterised by:
- the use of infrastructure intended to minimise latency (e.g. co-location, proximity hosting);
- system determination of order initiation, generation, routing or execution without human intervention for individual transactions or orders; and
- a large volume of messages throughout the day.
The authorisation requirement triggers other obligations such as requirements relating to risk management and capital adequacy as well as obligations to report to BaFin.
System and risk controls
Investment services enterprises engaging in algorithmic trading must have adequate system and risk controls in place for their trading systems (section 33 (1a) of the German Securities Trading Act (Wertpapierhandelsgesetz – WpHG). They must also have effective business continuity arrangements for handling unforeseen disruptions to their trading systems. Furthermore, they must ensure that any modifications to computer algorithms used in trading are documented appropriately. The requirement to maintain system and risk controls also applies to asset management companies and self-managed investment stock corporations.
Trading participants must designate orders generated through algorithmic trading and in doing so indicate the algorithm used (section 16 (2) no. 3 of the German Stock Exchange Act (Börsengesetz – BörsG). Exchange rules, which are subject to the approval of the exchange supervisory authorities of the federal states, contain specific provisions on identifying algorithm-based orders.
The Hessian Ministry of Economy, Transport and Regional Development (Hessisches Ministerium für Wirtschaft, Verkehr und Landesentwicklung), which acts as the exchange supervisory authority for the Frankfurt Stock Exchange, has published information on identifying trading algorithms.
Appropriate order-to-trade ratio
In addition to the above, trading participants must ensure an appropriate order-to-trade ratio in order to prevent risks to the proper functioning of exchange trading operations (section 26a of the BörsG). The various exchange rules contain specific provisions in this regard. Eurex has published circulars on the implementation of order-to-trade ratios.
Apart from the provisions referred to above, the Act imposes additional obligations on trading venues to charge separate fees for excessive use of exchange systems, to determine appropriate minimum tick sizes and activate circuit breakers to reduce volatility. The Act also clarifies that trading techniques involving computer algorithms may, under certain circumstances, also constitute market manipulation. Finally, the Act gives supervisory authorities powers to request information regarding algorithmic trading.
updated on 31.01.2014