Einde inhoudsopgave
EU Equity pre- and post-trade transparency regulation (LBF vol. 21) 2021/18.III.3.3.3
18.III.3.3.3 A specific type of algorithmic trading: high frequency trading
mr. J.E.C. Gulyás, datum 01-02-2021
- Datum
01-02-2021
- Auteur
mr. J.E.C. Gulyás
- JCDI
JCDI:ADS267147:1
- Vakgebied(en)
Financieel recht / Bank- en effectenrecht
Financieel recht / Europees financieel recht
Financiële dienstverlening / Financieel toezicht
Voetnoten
Voetnoten
See Netherlands Authority for the Financial Markets (AFM), A case analysis of critiques on high frequency trading, June 2016, p. 31.
See Netherlands Authority for the Financial Markets (AFM), A case analysis of critiques on high frequency trading, June 2016 and UK FCA (Financial Conduct Authority), Occasional Paper – Quantifying the High-Frequency Trading ‘Arms Race’: A Simple New Methodology and Estimates, January 2020 and ESMA, Consultation Paper: MiFID II/MiFIR review report on algorithmic trading, 18 December 2020(ESMA70-156-2368), p. 91-94 (ESMA discusses, among other things, the complexity in having so-called public and private feeds. Private feeds give a speed advantage to fast traders compared to investors that only have access to the public feed). For the sake of completeness, the ‘unfair’ speed advantage is contested by parties arguing that increases in speed are nothing new (e.g. the use of telescopes in the 17th century to see cargos on ships before the competition) and the same amount of speed is also available to institutional investors, as long as the investments are made. See Oliver Wyman, The Hidden Alpha in Equity Trading: Steps to increasing returns with the advanced use of information, 2013 (available at: https://www.oliverwyman.com/content/dam/oliver-wyman/v2/publications/2014/mar/The_Hidden_Alpha_in_Equity_Trading.pdfhttps://www.oliverwyman.com/content/dam/oliver-wyman/v2/publications/2014/mar/The_Hidden_Alpha_in_Equity_Trading.pdf).
Recital 62-63 MiFID II. A potential risk of high frequency trading is the commercial pressure to reduce the latency of the platforms. Each trading platform uses different methods to increase, which can result in a ‘race to the bottom’ and accordingly inconsistency in risk management practices (See Netherlands Authority for the Financial Markets (AFM), A case analysis of critiques on high frequency trading, June 2016, p. 33).
Ghost liquidity is the situation where a (high frequency) trader places duplicate limit orders on multiple competing venues, intending for only one of the orders to execute. Where one of the orders does execute, duplicates are cancelled. The rationale of this strategy is to reduce opportunity costs. Where a buy order for the same financial instrument is placed on multiple venue, there is more chance (opportunity) the order will be executed through a matching sell order. The risk of the ghost liquidity strategy is that the duplicate orders are both executed at the same time, thereby executing a too great quantity (See H. Degryse, R. De Winne, C. Gresse & R. Payne, ‘Cross-Venue Liquidity Provision: High Frequency Trading and Ghost Liquidity’,LIDAM Discussion Papers, 7 August 2019, p. 1).
See Netherlands Authority for the Financial Markets (AFM), A case analysis of critiques on high frequency trading, June 2016, p. 35.
ESMA, Economic Report: High-frequency trading activity in EU equity markets, 2014, p. 5.
See, for example, art. 48(1) (RMs) and art. 18(5) (MTFs) MiFID II. For an examination of the MiFID II framework for algorithmic trading, including high frequency trading, reference is made to D. Busch, MiFID II: regulating high frequency trading, other forms of algorithmic trading and direct electronic market access, Law and Financial Markets Review, 4 July 2016, p. 72-82. High frequency trading is typically done by investment firms trading only against their own capital (trading solely on own account), for example, as a market maker on an RM or MTF. Some high frequency traders have an SI status as well, which in essence means that client orders are executed through trading on own account on a large scale outside an RM or MTF (The Trade (H. McDowell), ‘The ELP SI Debate’, 30 October 2018 (available at: https://www.thetradenews.com/elp-si-debate/)).
ECB, Occasional Paper Series: Dark pools in European equity markets: emergence, competition and implications, July 2017, p. 3. Iceberg orders might not be sufficient here, because high frequency traders often rely on ‘pinging’ – sending small orders – to obtain information about hidden demand and supply on, for example, an RM or MTF. Once such an order is executed, a ping or series of pings alert the algorithm of the high frequency trader about the potential presence of a large order. For this reason, investors might want to protect order through completely dark options, such as the MiFID II large in scale-waiver (instead of the order management waiver). This is based on ECB, Occasional Paper Series: Dark pools in European equity markets: emergence, competition and implications, July 2017, p. 11 and p. 15.
One trading technique made possible through algorithmic trading is high frequency trading. High frequency trading is a subset of algorithmic trading, which is characterised by (a) high speed trades (very low latency) and (b) large amounts of transactions (hence, ‘high frequency trading’).1 The benefits and risks of high frequency trading are debatable. On the one hand, high frequency traders can be seen as a sort of arbitrage: abnormal prices are very quickly removed from the market. The benefit would be better/more efficient price formation.2 Other benefits of high frequency trading include increased liquidity through market-making strategies, narrower spreads, reduced short-term volatility and the means to obtain better execution of client orders.3 There are also concerns. High frequency trading can give an ‘unfair’ speed advantage in comparison to other market participants.4 Another worry is that – besides volatility5 and IT-related risks6 – high frequency traders can use strategies involving the quick cancellation of their orders, thereby displaying pre-trade data that is not executable in practice (‘ghost liquidity’).7
Two main effects of high frequency trading on equity pre- and post-trade transparency relate to (1) price formation and (2) dark liquidity. As noted, the effect of high frequency trading on price formation (point 1) is contentious. Opponents argue, among other things, that the fast speed of trading of high frequency trading does not contribute to price formation for equity pre-trade data due to ghost liquidity (i.e. orders are immediately cancelled).8 For this reason, one can conclude that post-trade data is more reliable than pre-trade data. The argument here is that the order is already executed and the related data (post-trade data) cannot be altered. This is not the common EU view, at least not of ESMA. ESMA accepts the proposition that high frequency trading helps to improve price formation, in particular through integrating trading activity on different venues.9 More broadly, MiFID II continues to permit high frequency trading, but introduces stricter rules, such as system control requirements.10 No distinct MiFID II equity pre- and post-trade transparency rules are in place for high frequency traders.
That being said, what is a well-accepted view is that high frequency trading resulted in a growing demand for dark liquidity (point 2). High frequency traders can prompt investors to choose to execute trades in venues where they can avoid interaction with high frequency traders.11 Under MiFID I high frequency trading resulted in a growing demand for dark liquidity, including large in scale-waivers and broker crossing networks, since dark liquidity can provide protection against the speed advantage of high frequency traders.12 The increase in dark liquidity resulted in a response of the EU in the form of restrictions on dark trading under MiFID II, in particular the double volume cap, stricter rules for SIs, and the intended reduction of trading outside RMs, MTFs, and SIs. Under MiFID II investors keep trying to find ways of avoiding high frequency traders. One way of doing so is apparent in the rise of ‘frequent batch auctions’ under MiFID II. Frequent batch auctions are examined below.