0

Faculty of Law

Academic Year 2017-18

Exam Session 1

High-Frequency Trading and Market Integrity

The Inadequacy of Market Structure following the Market

Abuse Regulation

LLM Paper

by Antoine MAIRESSE

Student number : 01711016

Promoter: Prof. Dr. Hans DE WULF

1

ACKNOLEDGEMENTS

I am thankful for the guidance provided by my promoter, Prof. Hans De Wulf, and his support

which allowed me to explore the fascinating, yet controversial, world of high-frequency trading.

I would like to acknowledge the professors which first raised my interest in financial law:

Xavier Dieux and Michèle Grégoire.

Writing this paper turned out to be a real journey trough the maze of capital markets. With its

ups and downs. It has shed a light on an ever-so-controversial part of our society and has further

raised my interest in the broader subject of innovations in financial markets.

I would also like to acknowledge my fellow LL.M. students. The quality of bonds that link us

brought the necessary motivation to see to the achievement of this paper and kept me going in

hardship.

“Restitute euis nomen universitatis in leatitia fraternitatis.”

2

INTRODUCTION

In 2014, Michael Lewis released an inflammatory book about High-Frequency Trading

(“HFT”) called Flash Boys – A Wall Street Revolt. Far from being derogatory in nature, it meant

to shed a light on the pre-eminence of HFT on the US stock market. Without going into details,

the book highlighted the particularities of the practice and concluded that the US equities market

was, somehow, rigged in favour of those able to trade at an ever-increasing speed. Of literary

nature, the narrative proposed by Lewis is obviously flawed from a scientific point of view and

was heavily questioned and debunked by the specialised press.1 However, it did put the subject

under the general spotlight and raised some valid and ongoing interrogations.

A few years before Lewis’ book release, HFT already made the headlines of the financial press

after the May 2010 “flash crash” where the Dow Jones Industrial Average dropped some 1000

points before sharply rebounding, in a matter of minutes, before the end of trading day. While

there is no certainty as to the role of HF traders in the crash, they were quickly blamed for it.2

After the trauma of the 2008 financial crisis, questions were once more raised on the capacity

of markets to support HFT activities.3

Thus, even though HFT is not a new practice, the general public’s interest in it is still quite

fresh. And logic has it that it is bound to keep stirring up interest for a long time. Especially

since technology and innovations keep driving the financial sector which is in constant

evolution. HFT is not immune to this evolution and, in the current markets structure, its

functioning is bound to improve. A 2010 declaration by the chief economist and the Executive

Director of Monetary Analysis and Statistics at the Bank of England illustrates this reality: “a

decade ago, the execution interval for HFTs was seconds. Advances in technology mean today’s

HFTs operate in milli- or microseconds. Tomorrow’s may operate in nanoseconds.”4

Nowadays, HFT has taken such a place on financial markets that regulators would have a hard

time turning a blind eye to the practice. And obviously, they have not remained inactive. The

rapid growth of HFT on European trading platforms warranted a response from the competent

authorities in the form of a recast of the Market in Financial Instruments Directive (“MiFID

1 See anthology of “opinions published in response to Michael Lewis’ Flash Boys and recent comments made by

NY Attorney General Eric Schneiderman”: FIA PTG, Debunking the Myths of High Frequency Trading, 2014,

available at www.fia.org. 2 J.A. BROGAARD, “High Frequency Trading and Its Impact on Market Quality”, Northwestern University.

Kellogg School of Management. Northwestern University School of Law, 2010, p. 3; 3 E. KRUDY, “Is the May 6 "flash crash" in US stock markets the new normal?”, Reuters, 10 June 2010, available

at www.reuters.com. 4 A. HALDANE, “Patience and Finance”, BIS Review, 114/2010, p. 10.

3

II”)5 and a brand-new Market Abuse Regulation (“MAR”)6. Among other goals, they intend to

regulate the risks arising from HFT7 and “provide measures regarding market manipulation that

are capable of being adapted to new forms of trading”8.

Those goals stem from the theory, on which most – if not all – financial regulations are based,

that to be efficient, markets need, among other things, to have a high degree of integrity to retain

investors’ confidence. Thus, abuses must be tackled for markets to function properly. However,

in our technology and innovation-driven society, identifying what is constitutive of an abuse

can be a daunting task. There lies the complexity of HFT.

At European level, it is important to assess whether the new legal framework is adequate to

tackle the challenges stemming from the rise of HFT on European markets and whether the

objectives set out in MAR – and indirectly MiFID II – regarding the prevention of market abuse

related to HFT are achievable with the current financial market structure.

The intention of this paper is therefore threefold. First, it will try to provide a general overlook

of HFT from a conceptual and legal perspective and build upon it to pinpoint the existing issues

posed by HFT. Second, it will try to propose a new paradigm through the lens of which market

integrity could be seen, namely the velocity paradigm, to replace the information paradigm in

accordance with which most financial legislation are adopted. Finally, it will attempt to assess

whether the current market design is still adequate to provide answers to the current challenges

face by modern markets.

In order to tackle these subjects, this paper shall be divided in three parts. The first part will

address the core concepts and principles essential to the comprehension of the problematic at

hand (Title I). Then, the second part will question the adequacy of the information paradigm by

concentrating on the issues pertaining to the apparent clash between HFT and market integrity

(Title II). And finally, the last part will be dedicated to examining how to resolve these issues

from a structural point of view. Leads and proposals on a regulatory level will be evaluated

(Title III).

5 Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial

instruments and amending Directive 2002/92/EC and Directive 2011/61/EU. 6 Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April 2014 on market abuse

(market abuse regulation) and repealing Directive 2003/6/EC of the European Parliament and of the Council and

Commission Directive. 7 Recital 59 of MiFID II. 8 Recital 38 of MAR.

4

TITLE I – GENERAL CONCEPTS AND PRINCIPLES

This title will be dedicated to set out the core concepts and principles which lie at the heart of

the subject. We will begin by introducing high-frequency trading in all its aspects in order to

provide an insight into its functioning and the underlying issues (Chapter I). And then, we will

focus on market integrity in order to understand how it can clash with HFT (Chapter II).

CHAPTER I – HIGH-FREQUENCY TRADING

In recent years, starting as far back as the 1970’s in the U.S.,9 direct human involvement in the

trading process has drastically diminished and machines have been taking over the world of

finance.10 Without going back to the whole modernization of stock exchanges11, we still need

to reflect first on the rise of algorithmic trading (section 1) in order to comprehend the

intricacies and functioning of HFT (section 2).

Section 1 – Technology and financial markets: a brief history of the rise of algorithmic

trading

While we still carry around the image of busy trading floors filled with brokers with gold

watches and hair slick with gel, screaming at their landline and making thousands or millions

with each phone call, from popular movies such as Wall Street and, more recently, Wolf of Wall

Street, the reality happens to be a lot less glamorous. To remain as efficient as possible, markets

have always been technology-driven. Indeed, “two significant interrelated technological

changes” have shaped the current markets: “investors using computers to automate their trading

processes and markets reorganizing themselves so virtually all markets are now electronic limit

order books.”12 But, admittedly, a room full of dashing young traders is more appealing than a

room full of computers.

The shift to electronic trading paved the way for algorithmic trading since “the speed and

quality of access to electronic markets encourages [its] use.”13 With algorithms, “recent years

have been marked by a shift towards a near-fully automated marketplace.”14 The algorithmic

trading craze trend has been prominent on US markets for more than a decade. However, no

9 Y. YADAV, “How Algorithmic Trading Undermines Efficiency in Capital Markets”, Vand. L. Rev., 2015, p.

1619. 10 A.P. CHABOUD et al., “Rise of the Machines: Algorithmic Trading in the Foreign Exchange Market”, J.

Finance, 2014, p. 2045. 11 For more on the subject, see A. ONOFREI, La négociation des instruments financiers : Au regard de la directive

MIF, Brussels: Larcier, 2012, pp. 149 and f. 12 T. HENDERSHOTT and R. RYORDAN, “Algorithmic Trading and Information”, Working Papers from NET

Institute, No 09-08, 2009, p. 2. 13 T. HENDERSHOTT and R. RYORDAN, Ibidem, p. 2. 14 Y. YADAV, “The Failure of Liability in Modern Markets”, Va. L. Rev., 2016, pp. 1034 to 1035.

5

time was lost by European markets to integrate this practice. In 2007, at the onset of the entry

into force of the brand-new Markets in Financial Instruments Directive (“MiFID”)15, trading

technologies kept evolving and automation started to make its way as a key placeholder in

European markets. “The resulting trading landscape can be characterised by higher competition

between trading venues, the fragmentation of trading in the same financial instruments across

venues in the EU as well as the increased use of fast and automated trading technologies.”16

Quickly outdated by those changes, MiFID was in dire need of a revision in order to address

the related issues. This recast was achieved in 2014 and aptly named MiFID II, which is the

seat of the regulation of trading venues at European level.

To define algorithmic trading is not such a difficult task. Some literature opted for a clear and

straightforward approach: “the use of computer algorithms to automatically make trading

decisions, submit orders, and manage those orders after submission.”17 And, for the purpose of

its legislation, the EU opted for a more technical definition in MiFID II: “trading in financial

instruments where a computer algorithm automatically determines individual parameters of

orders such as whether to initiate the order, the timing, price or quantity of the order or how to

manage the order after its submission, with limited or no human intervention, and does not

include any system that is only used for the purpose of routing orders to one or more trading

venues or for the processing of orders involving no determination of any trading parameters or

for the confirmation of orders or the post-trade processing of executed transactions.”18 While

both definition convey the same idea, the former makes for a clearer understanding of the

concept. As for algorithm, it can be defined as a pre-programmed electronic instruction, which

emphasises its ability to perform its tasks without relying on human intervention – outside its

programming.19

Section 2 – High-frequency trading: automation at its paroxysm

Now that we have dug into algorithmic trading, we must turn to high-frequency trading itself,

“a subset of algorithmic trading characterized by transactions executed at high volume and in

the space of milliseconds or microseconds.”20 Thus, first, we have to see how an activity is

15 Directive 2004/39/CE of the European Parliament and of the Council of 21 April 2004 on markets in financial

instruments amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European

Parliament and of the Council and repealing Council Directive 93/22/EEC. 16 ESMA Economic Report, “High frequency trading activity in EU equity markets”, Number 1, 2014, p. 5. 17 J.A. BROGAARD, Op. cit., p. 5 (definition originally from T. HENDERSHOTT et R. RYORDAN, Op. cit., p.

2). 18 Article 4.1 (39) of MiFID II. 19 Y. YADAV, Op. cit., p. 1035. 20 Y. YADAV, Op. cit., p. 1618.

6

identified as HFT (sub-section 1), before examining the typical HFT strategies (sub-section 2),

to finally understand how they are implemented and made feasible in practice (sub-section 3).

Sub-section 1 – Identifying high-frequency trading

The term “high-frequency trading” can refer to a number of groups of strategies that may, more

or less, vary depending on the person intent on giving a definition21, as the subject remains quite

controversial. However, it is unquestioned that, as a specific type of algorithmic trading, it is

characterised by extremely fast and high-volume transactions coupled with short holding

periods (seconds, minutes, hours but less than a day).22

In MiFID II, the EU provided a definition not for HFT itself, but for HFT techniques which it

defined as “algorithmic trading technique[s] characterised by:

(a) infrastructure intended to minimise network and other types of latencies, including

at least one of the following facilities for algorithmic order entry: co-location, proximity

hosting or high-speed direct electronic access;

(b) system-determination of order initiation, generation, routing or execution without

human intervention for individual trades or orders; and

(c) high message intraday rates which constitute orders, quotes or cancellations.”23

In 2016, a Delegated Regulation was adopted,24 and it provided a welcomed clarification as to

the meaning of “high message intraday rates,” which “shall consist of the submission on

average of any of the following:

(a) at least 2 messages per second with respect to any single financial instrument traded

on a trading venue;

(b) at least 4 messages per second with respect to all financial instruments traded on a

trading venue.”25

However, ever so technical, the EU still failed to provide a clear method to unambiguously

identify HFT activity in EU trading venues, which, admittedly, is not an easy task. In its 2014

21 For a non-exhaustive list of such definition, see I. ALDRIDGE, High-Frequency Trading: A Practical Guide to

Algorithmic Strategies and Trading Systems, 2nd edition, 2013, Hoboken, N.J.: Wiley, pp. 13 to 15. 22 I. ALDRIDGE, Ibidem, p. 15; J.A. BROGAARD, Op. cit., p. 5; Y. YADAV, Op. cit., p. 1618. 23 Article 4.1 (40) of MiFID II. 24 Commission Delegated Regulation (EU) 2017/565 of 25 April 2016 supplementing Directive 2014/65/EU of

the European Parliament and of the Council as regards organisational requirements and operating conditions for

investment firms and defined terms for the purposes of that Directive. 25 Article 19.1 of the Delegated Regulation.

7

economic report, the European Securities and Markets Authority (“ESMA”) observed two

methods taken from the literature to identify HFT activity, each with its merits and its faults:

(a) A direct approach which used a HFT flag based on the primary business of firms;

and

(b) An indirect approach based on the lifetime of orders which rely “on the ability of a

market participant to very quickly modify or cancel orders and can be computed for

individual stocks rather than at firm level.”26

The direct approach yielded lower results as it does not capture HFT activity by investment

banks.27 But despite the discrepancies, both approaches highlighted the same trends: HFT

accounts for roughly twice the number of orders than it does for the amount of value traded or

the number of shares.28

Overall, these definitions and methods of identification underline the essence of HFT: (1)

“physical proximity to exchanges; (2) rich data feeds of exchange activities; and (3)

programming that enables instant reaction to new information.”29

26 ESMA Economic Report, “Order duplication and liquidity measurement in EU equity markets”, Number 1,

2016, p. 12. 27 It stands to be noted that the report follows the indirect approach, as investment banks account for 61% of total

value traded, 22% of which is identified as HFT activity (ESMA Economic Report 2016, Ibidem, p. 13). 28 ESMA Economic Report 2016, Ibidem, p. 13. 29 Y. YADAV, “Insider Trading and Market Structure”, UCLA L. Rev., 2016, p. 995.

Figure 1 HFT activity in EU trading venues

Source: ESMA Economic Report, 2016

8

Sub-section 2 – High-frequency trading’s strategies

HF traders can implement an infinite amount of trading strategies with their algorithms.

However, besides traditional trading strategies, we can identify three broad classes of HFT

strategies based on their patterns and effects: arbitrage, market making and liquidity detection.30

First, arbitrage strategies take advantage of the algorithms capacity to “scour markets in search

of securities trading at a preferred price”31 and to detect price discrepancies of securities across

different trading venues.32 Less palatable to some, the latency arbitrage strategy takes advantage

of those discrepancies of price to quickly buy them at a lesser price in one venue to resell them

at a more advantageous price in another.33

In practice, a latency arbitrage strategy operation can be divided into these seven steps. Suppose

the security ABC:

“1. The market for ABC is $25.53 bid / offered at $25.54.

2. Due to latency arbitrage, an HFT computer knows that there is an order that in a

moment will move the [best bid and offer]34 quote higher, to $25.54 bid /offered at

$25.56.

3. The HFT speeds ahead, scraping dark and visible pools, buying all available ABC

shares at $25.54 and cheaper.

4. The institutional [algorithm] gets nothing done at $25.54 (as there is no stock

available at this price) and the market moves up to $25.54 bid / offered at $25.56 (as

anticipated by the HFT).

5. The HFT turns around and offers ABC at $25.55 or $25.56.

6. Because it is following a volume driven formula, the institutional [algorithm] is forced

to buy available shares from the HFT at $25.55 or $25.56.

30 I. ALDRIDGE, Op. cit., p. 15; Aldridge actually identify four broad classes. However, for the purpose of this

paper, arbitrage, market making and liquidity detection strategies suffice, as the second class, namely directional

event-based strategies, rely on the same principle as arbitrage in the context of specific events (see I. ALDRDIGE,

Ibidem, p. 148 and f.). 31 Y. YADAV, Op. cit., p. 1627. 32 I. ALDRIDGE, Op. cit., p. 16; Y. YADAV, Ibidem, p. 1627. 33 I. ALDRIDGE, Ibidem, pp. 196 and 197. 34 Refers to the “best advertised terms of trade immediately available for a security” (J. HASBROUCK, “The Best

Bid and Offer: A Short Note on Programs and Practices”, from the Finance Department, Stern School of Business,

14 October 2010, available at www.papers.ssrn.com, p. 2.).

9

7. The HFT makes $0.01-$0.02 per share at the expense of the institution.”35

As unpalatable as it is to some, by its input of demand and supply, it has the effect of re-

equilibrating the market price of the targeted security across venues. 36

Second, market making strategies rely on the ability and willingness of HF traders “to buy and

sell securities using their own money in order to keep the market liquid and orderly.”37 With

this kind of strategy, HF traders basically act as liquidity providers, usually at both end of the

spectrum – meaning that they post limit buy and limit sell orders. In MiFID II, an investment

firm is deemed to be pursuing a market making strategy “when, as a member or participant of

one or more trading venues, its strategy, when dealing on own account, involves posting firm,

simultaneous two-way quotes of comparable size and at competitive prices relating to one or

more financial instruments on a single trading venue or across different trading venues, with

the result of providing liquidity on a regular and frequent basis to the overall market.”38

HF traders, when acting as market makers play an important role by improving market liquidity

favour investors in that it greatly increases the chances for investors to find matches for their

sell or buy orders.39 Their role explain the specific attention given to such strategy in MiFID II.

Indeed, an algorithmic trader – not especially HF trader – must:

“(a) carry out this market making continuously during a specified proportion of the trading

venue’s trading hours, except under exceptional circumstances, with the result of providing

liquidity on a regular and predictable basis to the trading venue;

(b) enter into a binding written agreement with the trading venue which shall at least specify

the obligations of the investment firm in accordance with point (a); and

(c) have in place effective systems and controls to ensure that it fulfils its obligations under

the agreement referred to in point (b) at all times.”40

35 S. ARNUK and J. SALUZZI, “Latency Arbitrage: The Real Power Behind Predatory High Frequency Trading”,

Themis Trading LLC White Papers, 4 December 2009, p. 2. 36 I. ALDRIDGE, Op. cit., p. 197. 37 Y. YADAV, Op. cit., p. 1628. 38 Article 17.4 of MiFID II. 39 J.A. BROGAARD, Op. cit., p. 14. 40 Article 17.3 of MiFID II.

10

To function properly, they need to manage a sufficient inventory of the traded securities – too

big an inventory, they risk being unable to liquidate it, and too small an inventory is insufficient

to generate a profit – and keep track and respond to information.41

Finally, there are liquidity detection strategies. By far the most questionable class as it includes

an array of strategies which are outright abusive. They shall be discussed in Title I, chapter II,

section 2, sub-section 2 of this paper.

Sub-section 3 – High-frequency trading in practice

The characteristics of HFT uncovered in sub-section 1 are paramount to the proper functioning

of the strategies set up by HF traders. In practice, it is “through a combination of co-location,

direct feeds, and automated reactions to incoming information, [that] HFT traders can enjoy

first-sight and first-mover trading advantages.”42

Co-location is the uttermost expression of the physical proximity characteristic. It is paramount

for HF traders to access information as quick as possible. To do so, they started to locate their

servers as close as possible to the exchanges’ servers (e.g. in the building across the street),

which prompted exchanges to profit from this situation by selling space in the exact same

location. Co-location effectively ensures that information will reach the co-located servers of

HF traders before reaching anyone else.43 Co-location is usually combined with the purchase

of exchanges’ data feeds.44

However illustrative of HFT, co-location services are a consequence of HFT, not a cause.45 It

merely symbolizes HF traders’ need for speed. For instance, in 2010, Spread Networks

connected the New York and Chicago financial markets by an almost straight line of high-speed

fibre optic cables, at an estimated cost of $300 million. The speed of communication diminished

from 16 to 13 milli-seconds.46

Therefore, what proximity does is actually providing rich – direct in case of co-location – data

feeds which will be treated by algorithms capable of gathering, analysing and acting upon great

41 I. ALDRIDGE, Op. cit., p. 166. 42 Y. YADAV, Op. cit., p. 976. 43 Y. YADAV, Ibidem, p. 976. 44 Y. YADAV, Ibidem, p. 1016. 45 M. AITKEN, D. CUMMING and F. ZHAN, “Trade Size, High Frequency Trading, and Co-Location Around

the World”, European Journal of Finance, 2014, available at www.papers.ssrn.com, p. 20. 46 E. BUDISH, P. CRAMTON and J. SHIM, “The High-Frequency Trading Arms Race: Frequent Batch Auctions

as a Market Design Response”, Q. J. Econ., Vol. 130, 2015, p. 1549; The line has already been rendered obsolete

by the development of microwaves, because light – and therefore information – travels faster through air than

through glass (i.e. fibre optic cables). This emphasises further how reliant the financial sector is on innovation and

technology.

11

amount of data in record time. In the case of co-location, this means that by the time information

of the price of a given stock has reached non-co-located traders, HF traders have already acted

upon it and rendered that price obsolete.47 And while their profit margin on each trade is quite

small, their ability to trade in and out of positions, with tremendous velocity and consistency –

thousands, if not millions of times a day – and without holding positions overnight, makes up

for it.48

Professor Yadav “provides a [clear] schematic outline of how information is incorporated into

prices in a market with co-located HFTs:

Suppose that new information on Stock X is fed at Time 0 (simultaneously) into the direct feed

and the SIP feed49. Suppose also that Stock X is trading at $100 at Time 0, and the net effect

of this information will be to eventually change the price of Stock X to $101. This information

travels about 1000 feet along the direct feed to the co-located servers of HFTs and reaches them

at Time 1. HFTs algorithmically review and process the information and trade on it. The orders

are sent for execution back about 1000 feet to the exchange server and get executed at Time 2.

47 Y. YADAV, Op. cit., pp. 977 and 1001. 48 M. AITKEN, D. CUMMING and F. ZHAN, Op. cit., p. 4. 49 SIP feed refers to the Security Information Processor which collect, consolidate and disseminate market data.

“The SIP market data is used as a standard reference price for all U.S. securities trading” (Equity Markets

Association, “The Importance of Market Data to the U.S. Equity Markets”, EMA Whitepaper, 12 September 2012,

p. 2.

Figure 2 What happens with exchange information?

Source: UCLA L. Rev., 2016.

12

The trading actions of co-located HFTs changes the price to $101 at Time 2. All of this action

takes place potentially in the space of microseconds. The information travels along the direct

feed for (say) 100 miles and reaches other direct-feed investors at Time 3, well after the trading

actions of HFTs have already impacted the price at Time 2. The information also travels along

the SIP for (say) 100 miles and reaches other SIP investors at Time 4, typically after Time 3

and, in any case, well after Time 2.”50

This is where the competitive edge of HF traders lies.51 However, it is also the nest of the

controversial aspect of HFT.52

CHAPTER II – MARKET INTEGRITY: THE STRONGHOLD AGAINST

DISRUPTIVE HIGH-FREQUENCY TRADING PRACTICES?

Along with transparency, market integrity has been, and remains to this day, one of the two

governing concepts of the financial markets regulation. To grasp the importance of it, it must

be examined, first, through the scope of its objectives and underlying principles (section 1) and

then through the scope of its effective protection and, therefore, the prohibited behaviours and

violations of the current legal framework (section 2).

Section 1 – Market integrity

Traditionally, instead of providing a clear-cut definition for market integrity, regulators and

scholars alike focus on its goals and the means to achieve them through the prohibition of

numerous behaviours.53

But whatever the approach, assuring the integrity of securities markets is a worldwide concern

of financial regulations, as it is crucial for the economic welfare of nations.54 At the European

Union level, in its preamble, the new MAR, replacing the former Market Abuse Directive

(“MAD”)55 and applicable since 3 July 2016, essentially provides that market integrity is

paramount to guarantee the smooth functioning of and the investor’s confidence in the securities

markets and the financial instruments traded therein.56

50 Y. YADAV, Ibidem, p. 1033. 51 Y. YADAV, Ibidem., p. 976. 52 See infra, Title II. 53 For a complete oversee of the definition of (or the lack thereof) market integrity see J. AUSTEN, “What Exactly

Is Market Integrity: An Analysis of One of the Core Objectives of Securities Regulation”, Wm. & Mary Bus. L.

Rev., Vol. 8, 2017, pp. 215 and f. 54 J. AUSTEN, Ibidem, pp. 215, 218 and 236. 55 Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and

market manipulation (market abuse). 56 Recitals 4 and 10 of MAR.

13

On a securities market, it is essential for investor to have a certain level of trust on how the

assets are valued. Currently, it is mainly ensured through “disclosure regulation” – regulation

aiming at enhancing to a maximum the transparency of financial markets. A transparent market

is a reliable market in which investors are protected from misrepresentation of in the valuation

of assets.57 MAR, and by extension market integrity, through its prohibitions, acts as the strong

arm of securities markets regulation by ensuring that market players act with respect of those

principles.58

Therefore, the key idea behind promoting and actively safeguarding market integrity is to

maintain a certain level playing field between investors in order for them to be able to put their

trust in the functioning of the markets and therefore participate in them. In doing so, efficiency

is, up to a certain level, guaranteed.

Since market integrity is supposed to be achieved through the absence of market abuse, it is by

examining these unlawful behaviours constituting of market abuse that we shall comprehend

how HFT might jeopardize the integrity of financial markets.

Section 2 – Market abuse

To safeguard market integrity, MAR prohibits several behaviours in the financial markets

considered to constitute market abuse.59 Two of which must be examined, as they are

particularly relevant to the issues raised by high-frequency trading: insider dealing (sub-section

1) and market manipulation (sub-section 2).

Sub-section 1 – Insider dealing

The prohibition of insider dealing intended to protect investors. Preventing so-called “insiders”

from utilising inside information available to them prevents “normal” investors from

systematically losing to them. And it is the knowledge that they are not predestined to lose

against these insiders that drives them to participate in markets, therefore increasing the capital

available for investment. This is the system-wide benefit of the prohibition.60

57 J. AUSTEN, Op. cit., pp. 236 to 238; X. DIEUX, “Du « marché efficient » au « marché fiable » ? À propos de

la nouvelle législation belge en matière de transparence” in Droit, morale et marché, Brussels: Bruylant, 2013, p.

444. 58 It can be difficult to differentiate market integrity and transparency as they are complementary and often

intertwined since they seek to achieve grosso modo the same goals: investor protection and market efficiency – or

reliability according to some, see X. DIEUX, Op. cit., pp. 439 and f. 59 Recital 7 of MAR. 60 Y. YADAV, Op. cit., p. 972.

14

Accordingly, in MAR, the EU contends that “the essential characteristic of insider dealing

consists in an unfair advantage being obtained from inside information to the detriment of third

parties who are unaware of such information and, consequently, the undermining of the

integrity of financial markets and investor confidence.”61

Consequently, the Regulation provides that “a person shall not engage or attempt to engage in

insider dealing”62 and that “insider dealing arises where a person possesses inside information

and uses that information by acquiring or disposing of, for its own account or for the account

of a third party, directly or indirectly, financial instruments to which that information relates.

The use of inside information by cancelling or amending an order concerning a financial

instrument to which the information relates where the order was placed before the person

concerned possessed the inside information, shall also be considered to be insider dealing.”63

Hence, the prohibition is based on the concept of “inside information”, which is defined in

MAR as “information of a precise nature, which has not been made public, relating, directly or

indirectly, to one or more issuers or to one or more financial instruments, and which, if it were

made public, would be likely to have a significant effect on the prices of those financial

instruments or on the price of related derivative financial instruments.”64

From a conceptual point of view, it is not difficult to see how HFT clashes with insider dealing

as the algorithms are programmed to react and trade on the basis of not-fully-public

information.65 This issue shall be dealt with in Title II.

Sub-section 2 – Market manipulation

Article 12 of MAR deals quite extensively with what sorts of behaviours are constitutive of

market manipulation. Two types of market manipulation are highlighted by the legislation,

namely manipulations based on specific behaviours and manipulations based on information.66

The article is worded as follows:

“For the purposes of this Regulation, market manipulation shall comprise the following

activities:

(a) entering into a transaction, placing an order to trade or any other behaviour which:

61 Recital 23 of MAR. 62 Article 14 of MAR. 63 Article 8 of MAR. 64 Article 7 of MAR. 65 Y. YADAV, Op. cit., p. 977. 66 F. LEFÈVRE et al., “Les nouvelles règles en matière d’abus de marché”, R.D.C.-T.B.H., Vol. 5, 2017, p. 494.

15

(i) gives, or is likely to give, false or misleading signals as to the supply of,

demand for, or price of, a financial instrument, a related spot commodity contract

or an auctioned product based on emission allowances; or

(ii) secures, or is likely to secure, the price of one or several financial

instruments, a related spot commodity contract or an auctioned product based on

emission allowances at an abnormal or artificial level;

unless the person entering into a transaction, placing an order to trade or engaging in

any other behaviour establishes that such transaction, order or behaviour have been

carried out for legitimate reasons, and conform with an accepted market practice as

established in accordance with Article 13;

(b) entering into a transaction, placing an order to trade or any other activity or behaviour

which affects or is likely to affect the price of one or several financial instruments, a

related spot commodity contract or an auctioned product based on emission allowances,

which employs a fictitious device or any other form of deception or contrivance;

[…]”67

Furthermore, acknowledging the ever-increasing automation of markets, the European

Lawmaker has provided that the disposition regarding market manipulation need to be

adaptable to new form of trading such as HFT and provide non-exhaustive examples of such

abusive behaviour.68

Accordingly, the Regulation provides that “the placing of orders to a trading venue, including

any cancellation or modification thereof, by any available means of trading, including by

electronic means, such as algorithmic and high-frequency trading strategies, and which has one

of the effects referred to in paragraph 1(a) or (b), by:

(i) disrupting or delaying the functioning of the trading system of the trading venue or

being likely to do so;

(ii) making it more difficult for other persons to identify genuine orders on the trading

system of the trading venue or being likely to do so, including by entering orders which

result in the overloading or destabilisation of the order book; or

67 Article 12.1 of MAR. 68 Recital 38 of MAR.

16

(iii) creating or being likely to create a false or misleading signal about the supply of, or

demand for, or price of, a financial instrument, in particular by entering orders to initiate

or exacerbate a trend.”69

Together with the general prohibition of market manipulation of article 1470, this precision has

the effect of prohibiting abusive HFT strategies such as manipulative use of “quote stuffing”

(§1), “pinging” (§2), “spoofing” (§3) or “pump and dump” (§4) strategies.71 However, MAR

has provided a way to circumvent the prohibition with “accepted market practices” (§5).

§1 – Quote stuffing

Quote stuffing refers to a strategy which “involves submitting an extraordinarily large number

of orders followed by immediate cancellation in order to generate order congestion”72 and

effectively “clog” the networks and matching engines in order to slow down other traders73 or

slow down the “[transmission of] prices from an exchange to the SIP, thereby creating arbitrage

opportunities between markets.”74 However manipulative it may be, abusive quote stuffing is

very difficult to detect, especially in the context of HFT, and it is unclear whether HF traders

actually engage in such practice.75

§2 - Pinging

Pinging (and front running) refers to “an immediate-or-cancel order that can be used to search

for and access all types of undisplayed liquidity”76 and “if they detect a large order coming

through they may increase their trading activity. The result of such an action by the HF traders

would be to drive up the cost for the non-HFT market,”77 by “buying up the liquidity in the

[security] and selling it back at higher or lower prices (depending on if it was a buy or a sell

order).”78

69 Article 12.2 (c) of MAR. 70 “A person shall not engage in or attempt to engage in market manipulation." 71 This is a non-exhaustive list of examples of “unsound” HFT strategies from the “liquidity detection” class of

HFT strategies. For a more complete overview of such “unsound” strategies, see I. ALDRIDGE, Op. cit., chapter

12, pp. 195 to 208. 72 J. GAI, C. YAO and M. YE, “The Externalities of High Frequency Trading”, Working Paper, 15 March 2012,

available at www.papers.ssrn.com, p. 1. 73 I. ALDRIDGE, Op. cit., p. 201. 74 S. MCNAMARA, “The Law and Ethics of High-Frequency Trading”, Minn. J.L. Sci. & Tech., Vol. 71, 2016,

p. 116. 75 S. FRIEDERICH and R. PAYNE, “Order-to-trade ratios and market liquidity”, Journal of Banking and Finance,

2015, p. 215; J. GAI, C. YAO and M. YE, Op. cit., p. 3. 76 As per the definition provided by the US Securities and Exchange Commission (“SEC”) (J.A. BROGAARD,

Op. cit., p. 4. 77 J.A. BROGAARD, Ibidem, p. 21. 78 G. SCOPINO, “The (Questionable) Legality of High-Speed “Pinging” and “Front Running” in the Futures

Markets”, Conn. L. Rev., Vol. 47, 2015, p. 607.

17

§3 - Spoofing

Spoofing refers to a type of “momentum ignition strategy” that seeks to make other traders

believe that large trading interest is present in the market by introducing an order which is not

intended to be executed with the effect of distorting the order book.79 “And if prices react to

this phantom demand, HF traders can profit by trading into the reaction.”80 In doing, they

benefit from the artificial distortion they created.

§4 - Pump and dump

Pump and dump refers to a “classic” scheme whereby traders “pump” or artificially raise the

value of a particular financial instrument and then “dump” it at the first opportunity thereby

profiting from the artificial pike in the instrument’s value at the expense of other investors. This

particular type of market manipulation is not contested and has long been known by the

regulators. Such a scheme is a perfect illustration of information-based manipulation81 and is

not the prerogative of HF traders as it does not intrinsically rely on speed or data advantages.82

However, “HFT technology [can be used] to create certain conditions in the market, often [by]

manipulating other algorithmic trading programs, which the original HFT firms then takes

advantage of. It is as if HFT creates the weather in the financial ecosystem and then profits from

it.”83

§5 – Accepted market practices

While these four types of strategies have been identified as being manipulative, they remain

difficult to distinguish from strategies that are non-manipulative in nature and can result from

legitimate market practices.84 Accordingly, MAR contains an “accepted market practices”

provision which that a person engaging in activities constitutive of market manipulation may

do so “provided that the person […] establishes that such transaction, order or behaviour have

been carried out for legitimate reasons, and conform with an accepted market practice as

established […] by a competent authority [...] taking into account following the criteria:

79 I. ALDRIDGE, Op. cit., p. 202; C.R. KORSMO, “High-Frequency Trading: A Regulatory Strategy”, U. Rich.

L. Rev., Vol. 48, 2013, p. 548. 80 C.R. KORSMO, Ibidem, p. 548. 81 C.R. KORSMO, Ibidem, p. 554. 82 I. ALDRIDGE, Op. cit., pp. 202 and 203; S. MCNAMARA, Op. cit., p. 114. 83 S. MCNAMARA, Ibidem, p. 114. 84 I. ALDRIDGE, Op. cit., Chapter 12, pp. 195 to 208; T. ČUK and A; VAN WAEYENBERGE, “European Legal

Framework for Algorithmic and High Frequency Trading (Mifid 2 and MAR) A Global Approach to Managing

the Risks of the Modern Trading Paradigm”, Eur. J. Risk Regul., Vol. 9, 2018p. 152.

18

(a) whether the market practice provides for a substantial level of transparency to the

market;

(b) whether the market practice ensures a high degree of safeguards to the operation of

market forces and the proper interplay of the forces of supply and demand;

(c) whether the market practice has a positive impact on market liquidity and efficiency;

(d) whether the market practice takes into account the trading mechanism of the relevant

market and enables market participants to react properly and in a timely manner to the

new market situation created by that practice;

(e) whether the market practice does not create risks for the integrity of, directly or

indirectly, related markets, whether regulated or not, in the relevant financial instrument

within the Union;

(f) the outcome of any investigation of the relevant market practice by any competent

authority or by another authority, in particular whether the relevant market practice

infringed rules or regulations designed to prevent market abuse, or codes of conduct,

irrespective of whether it concerns the relevant market or directly or indirectly related

markets within the Union; and

(g) the structural characteristics of the relevant market, inter alia, whether it is regulated

or not, the types of financial instruments traded and the type of market participants,

including the extent of retail-investor participation in the relevant market.”85

With such a broad set of criteria, virtually any HFT strategy could fall within the scope of

accepted market practices, especially when it is already so difficult to discern between

legitimate and manipulative patterns of trading.

Nonetheless, such behaviours have already been the subject of prosecutions all around the

world. For instance, no later than 2014, in Japan, administrative monetary penalties have been

taken against an individual actively engaged in HF spoofing strategies.86 In the US, in 2011, a

company specialised in future and option contracts was also fined for the use of spoofing

tactics.87 And even more recently, three European banks – UBS, Deutsche Bank and HSBC –

paid hefty settlements to the for spoofing charges.88

85 Article 13 of MAR. 86 D. NIWA, “Market Manipulation Using High Frequency Trading and Issues Facing Japan”, Japan Lawyers

Guide, 2016/17, p. 39. 87 I. ALDRIDGE, Op. cit., p.202; Bunge Global Markets, Inc., CFTC Docket No. 11-10 (March 22, 2011). 88 M. PRICE, “European banks pay $46.6 million to settle U.S. 'spoofing' charges”, Reuters, 29 January 2018,

available at www.reuters.com.

19

TITLE II – FROM INFORMATION TO VELOCITY: A SHIFT IN FINANCIAL

MARKETS’ PARADIGM

In this part, we shall take on the challenges that have been highlighted in the previous part by

focusing on the fundamental change in the paradigm shaping the financial markets.

As previously stated, automated trading has taken over the financial market and with it, HFT

has gained tremendous foothold. At EU level, depending on the method used,89 HFT activity

accounts for 24% to 43% of value traded, 30% to 49% of the number of trades and between

58% to 76% of the number of orders.90 The rise was met with a new framework in the form of

a Directive and a Regulation: MiFID II with a “clear objective to instill more transparency

through harmonised EU rules on the financial market structure”91 and MAR safeguarding

integrity to ensure the essential transparency, “which is a prerequisite for trading for all

economic actors in integrated financial markets.”92

This emphasis on transparency is nothing new. And indeed, “information is an unusual

commodity. In financial markets, the right information can be worth millions,”93 which very

reminiscent of the infamous Gordon Gecko – arguably the most famous fictional trader –

teaching his young protégé that “the most valuable commodity [he] knows of is information,”

before being convicted of insider dealing in Oliver Stone’s Wall Street. This explain why “the

oldest means of investor protection is the provision of information”94 and why, therefore,

European legislation continue to press for more transparency.

However, the previous Title highlighted a growing asymmetry in the provision of information

when HF traders are involved due to the “never-ending and socially wasteful arms race” lead

by HF traders to gain speed in the transfer of data – and thus, information – to and from

exchanges and other trading venues.95

Therefore, based on the hypothesis that information paradigm through the lens of which modern

financial markets are designed is becoming outdated, we will review the requirements provided

in MiFID II and MAR regarding the practice of HFT (chapter I), then we will look into the

89 See Title I, section 1, sub-section 1. 90 ESMA Economic Report, Op. cit., 2016, pp. 12 and 13; For a diagram of those numbers, see figure 1 in Title I,

section 1, sub-section 1 taken from the same report. 91 T. ČUK and A. VAN WAEYENBERGE, Op. cit., p. 149. 92 Recitals 1 and 7 of MAR. 93 R.A. BREALEY and S.C. MYERS, Principles of Corporate Finance, New York: McGraw-Hill, 1991, p. 6. 94 V. COLAERT, “Recent trends in investor protection”, in Évolutions récentes en droit financier, under the

coordination of X. Dieux, Brussel: Bruylant, 2015, p. 51. 95 E. BUDISH, P. CRAMTON and J. SHIM, Op. cit., p. 1548.

20

market abuse rules. Precisely, we will examine the problematic posed by the structural

advantages enjoyed by HF traders challenging the traditional theory of insider dealing and

analyse the adequacy of the new market manipulation provisions when applied to HFT (chapter

II). Finally, we will conclude this Title by assessing the relevance of the information paradigm

and proposing another paradigm through the lens of which financial markets could – or should

– be structured.

CHAPTER I – MiFID II, MARKET INTREGRITY AND HIGH-FREQUENCY

TRADING: TRANSPARENCY AT THE RESCUE?

In 2012, in the aftermath of the Flash Crash and before the dawn of MiFID II and MAR,

Professor Wymeersch expressed “the need for efficient transparency in the markets.”96 And it

seems that his advice has been heeded. Transparency remains a cornerstone of financial

regulation. Which in itself is absolutely right. Overall, the benefits of transparency are not often

questioned.97

Besides the provisions already examined in Title I, the new framework provides an additional

disclosure requirement aimed specifically at investment firms partaking in HFT activity.

Indeed, MiFID II provides that investment firms that engage in HFT must “store in an approved

form accurate and time sequenced records of all its placed orders, including cancellations of

orders, executed orders and quotations on trading venues and […] make them available to the

competent authority upon request.”98

Such a provision, which specifically targets HFT, is only a small contribution to the grander

scheme set out to regulate all types of algorithmic trading, including HFT.99 This has the clear

96 E. WYMEERSCH, “Systemic Risk in the Securities Activities”, FLI UGent S&C, 2012-02, available at

www.law.ugent.be, p. 3. 97 Some scholars have questioned some rules based on transparency, such as insider dealing. According to them,

such rules exist at the expense of market quality (D.W. CARLTON and D.R. FISCHEL, “The Regulation of Insider

Trading”, Stan. L. Rev., Vol. 35, 1982, pp. 857 to 895). However, such calling into question of the legitimacy of

transparency is not predominant, in the literature or elsewhere. 98 Article 17.2, paragraph 5 of MiFID II. 99 Indeed, article 17 of MiFID II is very technical and provides a great deal of useful transparency to the supervision

and monitoring of the practice of algorithmic trading:

“1. An investment firm that engages in algorithmic trading shall have in place effective systems and risk controls

suitable to the business it operates to ensure that its trading systems are resilient and have sufficient capacity, are

subject to appropriate trading thresholds and limits and prevent the sending of erroneous orders or the systems

otherwise functioning in a way that may create or contribute to a disorderly market. Such a firm shall also have in

place effective systems and risk controls to ensure the trading systems cannot be used for any purpose that is

contrary to Regulation (EU) No 596/2014 or to the rules of a trading venue to which it is connected. The investment

firm shall have in place effective business continuity arrangements to deal with any failure of its trading systems

and shall ensure its systems are fully tested and properly monitored to ensure that they meet the requirements laid

down in this paragraph.

21

2. An investment firm that engages in algorithmic trading in a Member State shall notify this to the competent

authorities of its home Member State and of the trading venue at which the investment firm engages in algorithmic

trading as a member or participant of the trading venue.

The competent authority of the home Member State of the investment firm may require the investment firm to

provide, on a regular or ad-hoc basis, a description of the nature of its algorithmic trading strategies, details of the

trading parameters or limits to which the system is subject, the key compliance and risk controls that it has in place

to ensure the conditions laid down in paragraph 1 are satisfied and details of the testing of its systems. The

competent authority of the home Member State of the investment firm may, at any time, request further information

from an investment firm about its algorithmic trading and the systems used for that trading.

The competent authority of the home Member State of the investment firm shall, on the request of a competent

authority of a trading venue at which the investment firm as a member or participant of the trading venue is engaged

in algorithmic trading and without undue delay, communicate the information referred to in the second

subparagraph that it receives from the investment firm that engages in algorithmic trading.

The investment firm shall arrange for records to be kept in relation to the matters referred to in this paragraph and

shall ensure that those records be sufficient to enable its competent authority to monitor compliance with the

requirements of this Directive.

An investment firm that engages in a high-frequency algorithmic trading technique shall store in an approved form

accurate and time sequenced records of all its placed orders, including cancellations of orders, executed orders and

quotations on trading venues and shall make them available to the competent authority upon request.

3. An investment firm that engages in algorithmic trading to pursue a market making strategy shall, taking into

account the liquidity, scale and nature of the specific market and the characteristics of the instrument traded:

(a) carry out this market making continuously during a specified proportion of the trading venue’s trading hours,

except under exceptional circumstances, with the result of providing liquidity on a regular and predictable basis to

the trading venue;

(b) enter into a binding written agreement with the trading venue which shall at least specify the obligations of the

investment firm in accordance with point (a); and

(c) have in place effective systems and controls to ensure that it fulfils its obligations under the agreement referred

to in point (b) at all times.

4. For the purposes of this Article and of Article 48 of this Directive, an investment firm that engages in algorithmic

trading shall be considered to be pursuing a market making strategy when, as a member or participant of one or

more trading venues, its strategy, when dealing on own account, involves posting firm, simultaneous two-way

quotes of comparable size and at competitive prices relating to one or more financial instruments on a single

trading venue or across different trading venues, with the result of providing liquidity on a regular and frequent

basis to the overall market.

5. An investment firm that provides direct electronic access to a trading venue shall have in place effective systems

and controls which ensure a proper assessment and review of the suitability of clients using the service, that clients

using the service are prevented from exceeding appropriate pre-set trading and credit thresholds, that trading by

clients using the service is properly monitored and that appropriate risk controls prevent trading that may create

risks to the investment firm itself or that could create or contribute to a disorderly market or could be contrary to

Regulation (EU) No 596/2014 or the rules of the trading venue. Direct electronic access without such controls is

prohibited.

An investment firm that provides direct electronic access shall be responsible for ensuring that clients using that

service comply with the requirements of this Directive and the rules of the trading venue. The investment firm

shall monitor the transactions in order to identify infringements of those rules, disorderly trading conditions or

conduct that may involve market abuse and that is to be reported to the competent authority. The investment firm

shall ensure that there is a binding written agreement between the investment firm and the client regarding the

essential rights and obligations arising from the provision of the service and that under the agreement the

investment firm retains responsibility under this Directive.

An investment firm that provides direct electronic access to a trading venue shall notify the competent authorities

of its home Member State and of the trading venue at which the investment firm provides direct electronic access

accordingly.

The competent authority of the home Member State of the investment firm may require the investment firm to

provide, on a regular or ad-hoc basis, a description of the systems and controls referred to in first subparagraph

and evidence that those have been applied.

The competent authority of the home Member State of the investment firm shall, on the request of a competent

authority of a trading venue in relation to which the investment firm provides direct electronic access, communicate

without undue delay the information referred to in the fourth subparagraph that it receives from the investment

firm

[…]”

22

goal to provide the competent supervisory authority with sufficient data to assess the legality

of HFT activity. For instance, extreme episodic spikes in quoting activity are usually associated

with a degraded market quality.100 And if such occurrences are observed on markets, it can be

evidence of quote stuffing. However, identifying all quote-stuffing events is already a daunting

task,101 so identifying those quote stuffing events that unambiguously consist of market

manipulation is close to impossible.

Theretofore, transparency has been a useful tool in monitoring potential market abuse.

Unfortunately, it cannot be a cure for all ills and the aforementioned provision is close to moot

in the face of the sheer sophistication of the technology HF traders use. When taking into

account the difficulty to assess the abusive nature of behaviours adopted by HF traders,

transparency hardly seems to be an appropriate answer. Particularly, if we consider the

consistency with which financial technology have evolved and complexified, we have to accept

that HFT entails some sort of inherent opacity due to the ever-increasing sophistication of its

operating systems. Thus, requiring more transparency fails to address the elephant in the room.

In a 2014 speech, Martin Wheatly, formerly of the Financial Conduct Authority (“FCA”) in the

UK, expressed that “pattern detection alerts need to be able to detect suspicious behaviour

irrespective of the speed of trading. And these alerts are becoming ever more complex in order

to maintain the accuracy.” And that “the capacity of surveillance systems has to increase in line

with the trading technology and consequently the costs of surveillance rise too.”102 Which

alludes to the fact that to stay on top of effective monitoring, the FCA has to get in on the arms

race for better technology to keep up with HF traders, which further shows that providing more

data in not sufficient to properly monitor HFT if it is not correlated with a restructuration of the

market.

In a world of computer programming where the IT guy is now king, unconcerned by the legal

ramification of his highly technical work, transparency feels more and more like a band-aid on

an open wound.

The investment firm shall arrange for records to be kept in relation to the matters referred to in this paragraph and

shall ensure that those records be sufficient to enable its competent authority to monitor compliance with the

requirements of this Directive. 100 J. EGGINTON, B.F. VAN NESS and R.A. VAN NESS, “Quote Stuffing”, 22 March 2016, available at

www.papers.ssrn.com, p. 26. 101 J. EGGINTON, B.F. VAN NESS and R.A. VAN NESS, Ibidem, p.10. 102 M. WHEATLY, “Regulating high frequency trading”, speech made at the Global Exchange and Brokerage

Conference, New York, 4 June 2014, available at www.fca.org.uk.

23

CHAPTER II – MARKET ABUSE RULES: A SWING AND A MISS?

This chapter will consider the problems stemming from the nature of HFT – speed, proximity

and direct data feeds – and how it collides with rules on market abuse.

We will first study the concept of “structural insider dealing” which questions the illegal nature

of insider dealing by some market participants such as HF traders (section 1). And then, we will

examine the adequacy of market manipulation rules in the context of HFT (section 2).

Section 1 – Structural insider dealing

The concept of structural insider dealing has been thoroughly analysed in a recent study by

Professor Yadav.103 In her paper, she points out the inconsistencies in the coexistence of

traditional insider dealing theories and structural insider traders.

In order to comprehend this conflicting relationship, we will first study the general concept

(sub-section 1) and then see if and how it can be compatible with the European framework (sub-

section 2).

Sub-section 1 – General concept

The concept stems from the structural advantages – cutting-edge financial technology, physical

proximity and direct data feeds from trading venue – enjoyed by some market participants and

the fact that those high-speed structural insiders are essential to the proper functioning of our

modern financial markets.104 Indeed, with their market making strategies coupled with cross-

venue arbitrage strategies, both made possible by their ability to trade large volumes at record

speed, they dominate the price formation process and price discovery.105

It has been observed that the existence of those structural insiders, while bringing myriad of

benefits, also come at a cost, as the “harms arising from structural insider trading are remarkably

close in substance to those decried under the conventional theory of corporate insider

trading.”106

The purported benefits of HFT activity acting as insiders are that they bring a great deal of

liquidity to the markets,107 they contribute actively to the reduction of transaction costs in a

fragmented marketplace where trading venues compete for order flow and they increase the

103 Y. YADAV, Op. cit., pp. 968 to 1033. 104 Y. YADAV, Ibidem, p. 977. 105 M.P. LERCH, “Trading Activities”, in European Captial Markets Law, edited by R. Veil, Oxford: Hart

Publishing, 2nd Ed., 2017, p. 496; Y. YADAV, Ibidem, p. 977. 106 Y. YADAV, Ibidem, p. 978. 107 See supra. Title I, chapter I, section 2, sub-section 2.

24

speed and quality of the price formation process.108 Moreover, some argue that they reduce –

more or less significantly depending on the study109 – volatility110 while others argue the exact

opposite. Even their involvement in price formation is not uncontested,111 which goes to show

that where HFT is involved nothing is all black or all white, there are only grey areas.

However, the structural advantages enjoyed by HF traders can erode the confidence of other

investors against which the market feels rigged as they are not in a position to compete with the

speed at which HF traders gather, process and react to information. In this sense, non-HF

investors are bound to lose against those structural insiders.112

In a marketplace where investor protection is paramount, the greatest harms to it in the context

of HFT is the ability of HF traders to forecast markets by catching the first glimpse of trading

venues’ data which allows them the predict – more or less accurately – the direction in which

markets are headed. And for this short privileged period – a few seconds at least – HF traders

are able to read the market better, quicker and, and earlier than others.113 Moreover, it is

purported that they are also able to anticipate order flows which would allow HF traders to

understand how fundamentally informed traders are likely to transact enabling them to react

ahead of them and to take a small portion of the available profit on potentially very profitable

trade.114 However, in its latest report, the ESMA relied on the findings of a study of UK’s main

‘lit’115 venues which found no evidence that HF traders were able to anticipate near-

simultaneous order flow from non-HF traders, therefore making risk free profits. And when

they found patterns consistent with the anticipation of non-HF traders’ order flow for longer

periods of time (i.e. seconds), they “could not conclude whether this is because HFTs can in

fact anticipate the order flow or whether they are faster to react to new information.”116

108 M.P. LERCH, Op. cit., p. 496; Y. YADAV, Op. cit., p. 978. 109 For instance, Scholars found that while HFT did reduce volatility, it did so with no great significance, if it even

had any effect on it (J.A. BROGAARD, Op. cit., p. 60; T. HENDERSHOTT and R. RYORDAN, Op. cit., p. 5). 110 Volatility refers to the variations in trading prices of a given stock over a certain period of time. High volatility

means a lot of variations, while low volatility means that the price is more or less stable. Low volatility is preferable

in markets. 111 M.P. LERCH, Op. cit., pp. 492 and 496. 112 Y. YADAV, Op. cit., pp. 978 and 991. 113 Y. YADAV, Ibidem, p. 1014. 114 Y. YADAV, Ibidem, p. 1014. 115 By opposition to ‘dark’ venues, such as dark pools, not concerned by much of the transparency requirements.

‘Lit’ venues include traditional exchanges, multilateral trading facilities (“MTFs”) and organised trading facilities

(“OTF”). 116 ESMA Economic Report, Op. cit., p. 8, which cites M. AQUILINA and C. YSUSI, “Are high-frequency traders

anticipating the order flow? Cross-venue evidence from the UK market”, FCA Occasional Papers, no 16, 2016.

25

The effects of the erosion of investor protection and consequently confidence is all too real

because mistrusting investors, weary of the structural asymmetry, will have no incentive to

participate in markets which will inevitably lose in efficiency and quality.

Nonetheless, it has been accurately noted that “markets have undergone a sea change in how

they generate efficiencies. Rather than rely on a small cohort of institutions to maintain

liquidity, modern markets depend on a group of high-speed trading firms for volume, liquidity,

and investor participation. In fulfilling this market making function, HFT traders enjoy first

access to trading information that enables them to see market information first and to change

prices before information reaches the wider market.”117

What this entails is a blatant conceptual incompatibility between structural insider dealing and

the traditional prohibition of insider dealing, aimed at ensuring a certain level-playing field

between all investors by ensuring equal access to information and therefore equal opportunity

to trade on it. It is done by preventing market participants from trading on the basis of

undisclosed inside information.118 If they wish to trade on such information, they need first to

disclose it to the general public, therefore stripping them of the unfair advantage.

Sub-section 2 – Structural insider dealing under the EU financial framework

The evaluation of the portability of the concept of structural insider dealing to the European

financial market will depend on whether or not it falls within the current scope of insider dealing

rules. And the answer to that question depends on whether the information which HF traders

are able to access faster than other investors that have the necessary resources falls within the

scope of non-public, or inside, information under article 7 of MAR.119

From a conceptual point of view, and for all the reasons pertaining to its effects coupled with

the use of not-fully-public information,120 one less scrupulous towards HFT would be quick to

judge that it, in fact, falls within the scope of traditional insider dealing. However, if the solution

was that simple, HFT would not account for such a huge part of all trading taking place across

Europe.

117 Y. YADAV, Op. cit., pp. 1001 and 1002. 118 See supra. Title I, chapter 2, section 2, sub-section 1; It also stands to be noted that in French, inside information

is called “information privilégiée”, which literally translates to “privileged information”, therefore emphasising

the privileged position that people would hold had they been allowed to trade on it. 119 H. BOONEN, “High Frequency Trading, Electronic Frontrunning and Structural Insider Trading Under the EU

Market Abuse Regulation: Need for Reform?”, NYU J. of Law & Bus., 27 November 2017, available at

www.nyujlb.org. 120 Y. YADAV, Op. cit., p. 992.

26

The deciding factor in assessing the legality of HF strategies which rely on “preferential access

to order information, [is that] the orders are necessarily already present in the market before the

HF traders [are] able to detect them.121 [A] HF trader only identifies the orders of other market

participants after they actually have appeared in the market. As a consequence, HF traders are

not technically conducting illegal activity because they are not relying on inside information.

Notwithstanding the systemic informational advantages that HF traders enjoy by virtue of co-

location facilities and private data feeds, these practices do not fall under the prohibition of

insider trading.”122

The implication of that “technicality” is that the concept of structural insider dealing is

absolutely portable to the European markets. And that portability has some further regulatory

implications since, as of yet, HF traders falls on the good side of the law and are therefore

unconcerned with risks of insider dealing – as long as it does not involve the use of information

not yet present on the market – despite the harms it can do to market integrity. Therefore, the

current state of the law fails to address, at least, one aspect of the potential harmful nature of

the controversial practice.123

Section 2 – Detecting manipulative high-frequency trading strategies

As we have stated time and again, market manipulations are difficult to identify in the context

of HFT and Professor Korsmo noted that “there is some reason to fear that HFTs may be able

to solve the classic problem always faced by would-be manipulators—how to get out at a profit

before the manipulative effect evaporates. Information effects from trading are “likely to be

symmetrical—that is, any change in price caused by manipulative trades is likely to be offset

when the manipulative trades are unwound.” The sheer speed of HFT could allow them to

manipulate and exit the market before other traders are able to react. Furthermore, the huge

order volumes generated by HFT could potentially mask illicit layering activity, making

manipulative trading more difficult to detect.”124

121 This is reminiscent of the SEC’s Regulation National Market System (“Reg NMS”) which provides that

exchanges must make sure that data is sent to the SIP and direct feeds at the same time. But data does not have to

reach the feeds at the same time, allowing HF traders to maintain their competitive edge through latency arbitrage.

Contravention to this requirement will result in fines to contravening exchange, as it happened to the New York

Stock Exchange (“NYSE”) in 2012 (G. STONE, “SIP vs. Direct Feeds Latency – What are the Rules?”, Bloomberg

White Paper, 15 May 2014, available at www.bloomberg.com). Both in the US and the EU, the insiders’ advantage

is not unlawful in nature but would become so if the data/information was not first made available to the public. 122 H. BOONEN, Op. cit. 123 Title III will address the regulatory challenge faced by the European Lawmaker in relation to structural insider

dealing. 124 C.R. KORSMO, Op. cit., p. 556.

27

However, Professor Korsmo contends that in practice, it appears that those same features that

facilitate potentially consequence-free manipulations on the part of HF traders are the very

reasons that, in case of a true HFT manipulation, actual harms to investors are largely limited

to traders who also use extremely sophisticated algorithms, as they are the only one able to react

quickly enough to be fooled by them. In the meantime, “normal” or low-frequency investors

which trade during the period of manipulation also stand to be harmed. “But so long as such

trading is unrelated to the manipulation, it will be functionally random—the unknowing

investor is as likely to benefit as to suffer from any given manipulation.”125 You can only thrive

for so much investor protection since financial trading is a speculative activity that comes with

its fair share of inherent risks.

This is important because, even the Commission diligently included a non-exhaustive list of

prohibited HFT practices to exemplify the most common manipulative HFT strategies.

However, probably owing to the fact that HFT manipulations are just high-tech version of

traditional market manipulation,126 the Commission seems not to have taken advantage of the

recasting of the market abuse rules to questions their adequacy to tackle potential HFT

manipulations. Once more, they acknowledged the shift to new forms of trading, such as

HFT,127 but they failed to grasp the essence of this shift. The difficulty with which HFT

strategies can be unambiguously identified as manipulative stems not from a lack of information

or data to analyse. It stems from the velocity – itself a consequence of the core features of HFT:

proprietary trading, very short holding periods, submission of a large number of orders that are

cancelled shortly after submission, neutral positions at the end of a trading day and use of

colocation and proximity services to minimise latency –128 at which HF traders operates.

The only real novelty is in the monitoring department by imposing compliance requirements

on investment firms129 and trading venues.130 Accordingly, “control will be imposed on several

levels of the life cycle of orders, from the investment firm, to the direct electronic market access

provider (DEA), and the trading venue. […] Algorithmic orders will be flagged and “unusual

orders” must be blocked before being sent to the venue by pre-trade controls with predetermined

thresholds in terms of prices and volumes. When trading activity seems unusual, the firm and

the venue should be able to either stop an algorithm with a kill function or stop overall trading

125 C.R. KORSMO, Op. cit., p. 557. 126 S. MCNAMARA, Op. cit., p. 71. 127 Recital 38 of MAR. 128 ESMA Economic Report, Op. cit., 2016, p. 5. 129 Article 17 of MAR. 130 Article 48 of MAR.

28

in a given instrument by a circuit breaker.”131 But from the point of view of manipulation

detection, improving the monitoring of algorithm is pointless if the distinction between truly

disruptive behaviours and legitimate practice remains too complicated to draw.

Nonetheless, whatever the shortcomings of the market manipulation rules as provided for in the

new European framework, one should not be to harsh in assessing the adequacy of the

concerned measures since they are as adequate as they can be with the systemic and principles-

based approach chosen by the Commission.

In conclusion, it seems that despites all the efforts made by the European Lawmaker to provide

rules on market manipulations inclusive of issues related to HFT, it appears unlikely that they

will have tangible effects in practice. Indeed, even though HFT is, step by step, becoming the

a norm in financial market and was therefore in dire need of a regulatory framework - duly

provided by MiFID II, trying to address its potential abusive nature with the traditional and

unrevised notions of market abuse – market manipulation and insider dealing alike - coupled

more transparency is thoroughly inefficient. Efficient rules to prevent market manipulation

must take into account the specific nature of HFT,132 which could prevent regulatory

inconsistencies such as that “while algorithms operate with precise quantitative data, the

prohibited effects are expressed in broad qualitative terms”133 or uncertainties linked to the

complexity of HFT strategies making it unclear under which prohibition they may fall.134

CHAPTER III – MARKET INTEGRITY THROUGH THE LENS OF THE VELOCITY

PARADIGM

In the two previous chapter, we have examined the approach chosen by the Commission to

maintain market integrity in markets more and more dominated by HF traders. That is the

“insanity” approach: doing the exact same thing over and over again and expecting a different

outcome.

While this is an obvious hyperbole, it remains that for MAR, the Commission merely beefed

up the provisions of MAD135 in order to encompass HFT activity but without giving more

thoughts to their ability to actually prevent disruptive behaviours such as insider dealing and

131 T. ČUK and A. VAN WAEYENBERGE, Op. cit., p. 151. 132 Title III will address regulatory measures emphasising on the characteristics of HFT to ensure integrity. 133 T. ČUK and A. VAN WAEYENBERGE, Op. cit.., p. 153. 134 T. ČUK and A. VAN WAEYENBERGE, Ibidem, p. 153. 135 For the provisions on insider dealing and market manipulation, see article 3 and 5 of MAD.

29

market manipulation. It begs the question why? Especially since they were on the verge of

drafting a tremendously ambitious recast of the framework for European financial markets.

The failure to take into account the specificity of HFT when market integrity and market abuse

were concerned is even more dubious when we consider that four years prior to the adoption of

MiFID II and MAR, the 6 May 2010 Flash Crash made the headlines all over the world. Indeed,

in the aftermath of the Flash Crash, market participants and regulators alike were quick to blame

HFT. The US Department of Justice and the Commodity Futures Trading Commission

(“CTFC”) investigated the anomaly which shook the trading world for less than an hour.

The CTFC investigation found that HF traders did not cause the Flash Crash but contributed to

its depth. And a year later, the US Department of Justice arrested Navinder Singh Sarao, a

British stock brokers, for his active role in the event of the 6 May 2010. Apparently, Sarao had

been involved in spoofing for a few years. He “[created] a strong ceiling in the market in the

form of many or very large sell orders just under the best available price. As these orders were

visible to other traders, they created the illusion that prices were about to fall. This caused

downward pressure on the market, which prompted an actual fall in prices. Sarao then bought

in cheaply. Afterwards he cancelled the original sell orders, thereby removing the ceiling and

allowing prices to rise again.”136 And according to the CTFC, the Flash Crash was caused by

particularly aggressive spoofing tactics on his part. However, commentators were warier to

blame the whole failure to a single trader.137

This whole story is symptomatic of the grander problem faced by regulators. The inability to

detect unambiguously the cause of such a massive market failure was and still is traumatic and

extremely detrimental to investor confidence in the market. But the fact that the observation

made by Professor Menkveld at the time of the arrest of Sarao, namely that “those kinds of

[manipulative] tactics are extremely difficult to spot” – indeed, Saroa had been taking part in

spoofing strategies for at least 400 trading days over the years and only the significance of the

Flash Crash seems to have been able to expose him. “And if it can happen in America, it can

happen here,”138 is still true after the entry into force of the new European framework is even

more worrisome.

136 D. BUSCH, Op. cit., p. 72. 137 J.A. BROGAARD, Op. cit., p. 3; D. BUSCH, Ibidem, p. 72; E. KRUDY, Op. cit., available at www.reuters.com. 138 A. MENKVELD quoted in “Manipulatie in S&P-futures voorspelt weinig goeds voor andere markten”,

Financieele Dagblad, April 2015, available at www.fd.nl.

30

We should not have to wait for further incidents, failures or irregularities to have efficient HFT

market abuse rules. But supposing that new HFT market abuse rules have to be drafted, we

contend that it would have to come with a shift in paradigm.

Up until now, European financial regulations have been drawn up in accordance with the

“information paradigm”. Hence, financial regulations have long been following a common

principle-based pattern keen on transparency and information disclosure.139 MiFID II and MAR

are no exceptions to this pattern, which was best illustrated by Louis Brandeis in 2014 when he

wrote that “the sunlight is said to be the best of disinfectants.”140

However, it appears that nowadays, orders travel faster than sunlight. This is the reason why

we advocate for the application of a new paradigm through the prism of which financial

regulation pertaining to market integrity can be designed when high-speed technology such as

HFT are concerned: the velocity paradigm.

Obviously, this proposal does not negate the benefits of the information paradigm which

remains essential to the proper functioning of financial markets. It simply considers that this

paradigm is highly inadequate for the safeguarding of market integrity in the face of the ever-

increasing sophistication and complexification of HFT.

The “velocity paradigm” requires that, when appropriate, the emphasis be placed on the speed

at which information is gathered, processed and acted upon instead of the information itself,

since no one is to say when technological innovations will stop, if ever. Therefore, if speed is

not at the forefront of some regulatory rules, we face the risks of having HF traders being always

one step ahead of the regulation. And in the modern market, what is to stop them from

continuously gain advantages through ever-increasing velocity? This the reason why we find

the velocity paradigm to be better suited to ensure effective market integrity in the context of

HFT.

Accordingly, in the following Title, the critical analysis of the current market structure will be

done through the prism of the velocity paradigm. Meaning that we will assess whether the

measures effectively focus on speed in a manner that has positive effects on market integrity.

139 V. COLAERT, Op. cit. p. 92. 140 L. BRANDEIS, Other people’s money, and how the bankers use it, New York: Stokes, 1914, p. 92.

31

TITLE III – MARKET DESIGN IN MOVEMENT: THE ADEQUATE RESPONSE TO

HIGH-FREQUENCY TRADING?

This Title will concentrate on reconciling the issues raised in the previous titles with the goal

of market abuse rules.

We have observed incompatibilities between the current market design and the safeguarding of

market integrity in the context of HFT. And our observations have lead us to agree with the

statement that “the high-frequency trading arms race is a symptom of flawed market design.”141

Indeed, we have seen that in the search for market integrity, the European Lawmaker has missed

the opportunity offered to him by simply incorporating HFT as any other kind of novel trading

tactic. Nonetheless, we have already found that HFT did not fit within the current market design

when it came to the prevention of market abuse. Therefore, the inadequacy between HFT and

market design has left HF traders in some sort of legal no man’s land where they are virtually

untouchable and all powerful. Almost nothing can effectively hinder their structural advantage,

outside of obvious or traditional abuses.

We will be to review proposals which, unlike the most recent legislation, accounts for the

advancement in HFT. As a practice that conceptually goes against the core principles of market

integrity, we will examine if and how HFT can be dealt with in manners consistent with both

the goals of market integrity and the purported benefits of HFT activity.

We will begin by assessing the remaining courses of action of policymakers within the current

market structure (chapter I). And then, we will assess how the prevention of HFT-related market

abuse can be enhanced, or simply made effective, by rethinking market design (chapter II).

CHAPTER I – RESPONSES TO HIGH-FREQUENCY TRADING UNDER THE

CURRENT MARKET STRUCTURE

Within the current market structure, the die is already cast and not much possibilities are left to

further regulate HFT in order to prevent the negative effect it can have on market integrity.

Two types action can be identified to safeguard market integrity: a ban of HFT (section 1), the

imposition of levies on HFT activity (section 2) or even drafting liability rules as an ex post

regulatory strategy.142

141 E. BUDISH, P. CRAMTON and J. SHIM, Op. cit., p. 1547. 142 C.R. KORSMO, Op. cit., p. 591; Under the new framework, “responsibility for disruptive events can now be

attributed to all relevant persons at all hierarchical levels, from the developer to the trader, the trading supervisor,

as well as compliance and the firm’s management” (T. ČUK and A. VAN WAEYENBERGE, Op. cit., p. 151).

32

Section 1 – Ban on high-frequency trading

A State may adopt an outright ban on HFT. While it is, theoretically, possible, it would simply

negate the evolution of modern markets. It would also negate all the benefits stemming from

HFT – liquidity provision, reduction of transaction costs. It would simply be too costly for all

market participants. At this point, it is not a realistic scenario anymore. And the same goes for

an EU-wide ban, even though the fickle and reactive nature of financial regulation143 may prove

us wrong in the hypothesis of further scandals and irregularities.144

Section 2 – Taxation of high-frequency trading activity: the Italian case

A lot less drastic than a ban, A State may decide to tax HFT activity in order to safeguard market

integrity. At EU level, States remains sovereign when it comes to taxation. And accordingly, in

2013, Italy became the first country to impose a tax on HFT by adopting the Italian Financial

Transaction Tax (“IFTT”). “The tax was reportedly introduced due to concerns that the growth

of HFT in Italy could potentially have an adverse impact on the integrity and quality of Italian

financial markets, particularly with regard to volatility and liquidity.”145

A State intent on levying taxes on HFT activity would be the provision a clear definition and

scope of the concept of HFT activity in order to have a predictable tax base. And we have

already observed the existing difficulties of identifying and defining HFT. However, the

discretionary powers of the State to provide its own definition for tax purposes would greatly

reduce the complexity of the task.

Indeed, a State willing to expend to the maximum its tax base would opt for a broad definition

with the effect of overestimating HFT activity, thus encompassing in its scope non-HFT

activity, with the risk of having “victims” of the overestimation raising concern and

jeopardizing the validity of the tax. And at the other end of the spectrum, a State may also opt

for a narrow definition which would likely cause underestimation on provide a lesser tax

base.146 For the IFTT, Italy opted relatively neutral scope – not too broad, not too narrow – that

appears to be in accordance with the goal of the tax and based on the algorithmic and quick

This regulatory strategy will not be analysed in this paper as it is a whole other issue in its own right which would

require its own study. For such a study, see Y. YADAV, Op. cit., Va. L. Rev., pp. 1031 to 1100. 143 Y. DE CORDT and G. SCHAEKEN WILLEMAERS, “Le concept de transparence en droit financier”, Rev.

prat. soc., 2006, p. 115. 144 D. BUSCH, “MiFID II: regulating high frequency trading, other forms of algorithmic trading and direct

electronic market access”, L.F.M.R., Vol. 10, 2016, p. 79. 145 R.S. MILLER and G. SHORTER, “High-Frequency Trading: Background, Concerns, and Regulatory

Developments”, CRS Report, 19 June 2014, p. 36. 146 The problematic of over- or underestimation of HFT activity was also raised by the ESMA in examining existing

methods of identification of HFT activity (ESMA Economic Report, Op. cit., 2014, p. 6.

33

nature of the transactions147: “there shall be deemed to be high-frequency trading those

transactions which jointly have the following features:

(a) they are generated by a computer algorithm that automatically determines the decisions

relating to the sending, modification and cancellation of orders and of the relevant parameters

[…];

(b) they occur at intervals not exceeding half a second. This interval is calculated as the time

between the placing of an order for purchase or sale and the subsequent modification or

cancellation of the same order by the same algorithm.”148

However, whatever the choice made by the State, it remains bound to some extent by the

essential role played by HF traders. De-incentivising to much HF traders could come at a price

for the financial markets of the State. That is why Italy choose to exempt market making

activities from its tax on HF transactions.149

And in the case of the IFTT, market participants raised concerns regarding the negative effect

such a tax would have on liquidity and early reports following the adoption of the IFTT

appeared to confirm the doubts as liquidity seemed to plummet quickly after the adoption of

the tax.150

With such a tax, the State aims to address the purported harms of many of HFT strategies

without addressing their legality and legitimacy. It both addresses and avoids the underlying

issues.

Therefore, while the adoption of a tax on HFT is a legitimate response from States in order to

safeguard integrity in the national markets, it seems to simply negate the benefits brought forth

by HFT. And while State interests are not intrinsically meant to align with those of the market,

levying taxes – therefore de-incentivising HF traders – is a bold move in such a fragmented and

competitive marketplace.

Market participants can prove to be great innovators in the battle to ensure market integrity

against HF traders.

147 R.S. MILLER and G. SHORTER, Op. cit., p. 36. 148 Article 12 of the Law Decree of the Italian Minister of Economy and Finance of 21 February 2013 (Decreto

del ministero dell'Economia e delle Finanze del 21 febbraio 2013), London Stock Exchange unofficial translation. 149 Clifford Chance, “The Italian Financial Transaction Tax Factsheet”, briefing note, January 2013, p. 1. 150 M. CLINCH, “Italy launches tax on high-frequency transactions”, CNBC, 2 September 2013, available at

www.cnbc.com.

34

CHAPTER II – RETHINKING MARKET STRUCTURE

We have extensively expressed the existence of purported benefits and harms of HFT related

to market quality and/or efficiency: increase or decrease of transaction costs, increase or

decrease of liquidity and increase or decrease volatility. Each study draws its own

conclusions.151 However, when it regards market integrity, even though studies do not really

agree on the negative impact of HFT, none argues that HFT bring about an increase in market

integrity.

Nonetheless, with the European framework, HF traders are more or less able to engage in HF

activity without much hindrance. The question is therefore to know what kind of structural

changes could be envisaged in order to restore investor confidence – into the financial markets

dominated by HF trader – and subsequently, safeguard market integrity.

We will examine two sorts of solutions: the first one will have us address the tacit acceptance

of structural insider dealing (section 1) and the second one will examine the possibility of

restoring a true level-playing field among investors (section 2).

Section 1 – Recognition of structural insider dealing: taking a stance

When contemplating regulating structural insider dealing, “the fundamental question is whether

differential access to exchange data creates harms for market quality, broadly understood. By

the conventional account, theory paints a bleak picture. From the viewpoint of investor

protection, the harms can be substantial and far-reaching. If insiders repeatedly get the best

deals, then other investors have little incentive to remain on the market, or to deploy their capital

fully. Markets are eventually drained of their power to allocate capital”152

Through the lens of the information paradigm, structural insider dealing is perfectly fine as the

information is made available simultaneously to all investors. It does not fall on the regulators

to control the time it takes to actually reach different investors.153 And we find ourselves in this

incoherent, though current, status quo which agrees that the harms of structural insider dealing

are essentially the same as the one from traditional corporate insider dealing but at the same

time, it does not fall within the scope of prohibited insider dealing.154

However, if you look at structural insider dealing through the lens of the velocity paradigm,

you have to consider the speed at which the information is processed. This way, insider dealing

152 Y. YADAV, Op. cit., p. 1018. 153 Much like in the US (G. STONE, Op. cit., available at www.bloomberg.com). 154 H. BOONEN, Op. cit., available at www.nyujln.org.

35

rules could not shy away from taking a stance on whether structural insider dealing falls within

the scope of the prohibition because overall, the debate boils to whether structural insider

dealing should be held as falling within the scope of conventional insider dealing or whether it

should remain outside the scope of insider dealing. One way or another, there is a need to find

coherence between the doctrine and policy.155 Indeed, “in case MAR should only be applicable

to core corporate insider trading, it is recommended to adjust the doctrine to reflect its limited

scope.”156 And if the contrary is preferred, it must be coupled with effective market

restructuration that do not leave structural insider dealing in a unregulated status quo.157

But whatever approach is chosen, it will be a stance since, as Professor Yadav expressed, if “the

law applies to one set of insiders to control harms [that are linked with a reduction in HFT

activity] points to a legal regime sorely out-paced by innovation and unable to consistently fulfil

its basic function. The rise of structural insider trading, fundamentally, calls into question the

inner coherence of the prohibition and points to the need to rethink its foundations. If the law

can only apply effectively to core corporate insiders, this should be made explicit and the

doctrine should be recalibrated to reflect this limited reach. But even this narrower application

poses a problem. Why should the law protect investors from one set of insiders, but leave them

open to harm from another? If it is ultimately a question of safeguarding investor protection,

there does not appear to be any good answer to the inquiry.”158

Section 2 – Restoration of the level-playing field

Having a level-playing field is a key component of ensuring market integrity.159 The concept

of “level-playing” relates to an ideal in which each investor is armed with the same amount of

information as the others allowing each and everyone of them to make informed investment

decisions.160 However, “[asymmetrical] access to information between investors is almost an

inherent characteristic of traditional markets.”161 And therefore, the only way to ensure some

sort of level-playing flied is through regulation. Prohibition of insider dealing is the best

illustration of how policymakers have tried to ensure this level-playing field.

However, as we have seen previously, “in practice HFT calls into question the notion of a level

playing field by restricting the ability to profit from it to only the very most sophisticated

155 H. BOONEN, Ibidem; Y. YADAV, Op. cit., p. 1032. 156 H. BOONEN, Ibidem. 157 Such measure will be examined in section 2 158 Y. YADAV, Op. cit., p. 1032. 159 J. AUSTEN, Op. cit., p. 235. 160 S. MCNAMARA, Op. cit., p. 145. 161 H. BOONEN, Op. cit., available at www.nyjlb.com.

36

actors.”162 Indeed, “traders with speed advantages will always win the competition to come in

first, as others can never make up for any initial delay.”163

This where reshaping market structure comes into play. If investors cannot make up for the

delay themselves, market structures can. Therefore, we will analyse the equalisation of access

as it has been implemented (sub-section 1) and then we will analyse an alternative to the

continuous-time trading model, which is the one used on most trading venues164 (sub-section

2). Both alternative structures tackle head-on the speed component of the structural advantage

enjoyed by HF traders.

Sub-section 1 – Equalising access: the case of IEX

The modern marketplace is extremely fragmented between all sorts of trading venues. And

fragmentation means an increase in competition. Nowadays, market operators need a

competitive edge almost as much as the investment firms and other market participants. It

usually translates to fee reduction on the part of trading venues, which is also a consequence of

algorithmic trading. Indeed, the Deutsch Boerse observed that “algorithm generated trading is

lower cost and highly sensitive to fee reductions and therefore, could receive quantity

discounts.”165

Another way to be competitive is to tailor your venue’s structure to offer innovative services.

That was the path taken by the stars of Michael Lewis’ Flash Boys and founders of IEX (the

“investors’ exchange”). In order to circumvent the structural advantages of HF traders, they

structured their exchange around a 350 microseconds built-in delay on incoming orders which

allows a wider range of algorithmic traders to compete for orders.166

With built-in delays, such as the one of IEX,167 “outsiders may be better able to transact on up-

to-date information. They may be able to at least see the latest prices, before HFTs can submit

orders to trade on them and update prices,” which “help reduce the disparity in access to price

data”168 and prevent outsiders from trading on outdated information.

162 S. MCNAMARA, Op. cit., p. 145. 163 S. MCNAMARA, Ibidem, p. 144. 164 E. BUDISH, P. CRAMTON and J. SHIM, Op. cit., p. 1549. 165 A. CHABOUD et al., Op. cit., p. 2. 166 Y. YADAV, Op. cit., pp. 1028 and 1029. 167 IEX is not alone in this endeavour. TMX also has a build-in “speed bump,” which is a random-length delay of

1 to 10 milliseconds. This is based on the idea “that millisecond-level randomness dwarfs any microsecond-level

differences in speed among trading firms responding to the same stimulus, which in turn reduces the incentive to

invest in tiny speed improvements” (E. BUDISH, P. CRAMTON and J. SHIM, Op. cit., pp. 1610 and 1611). 168 Y. YADAV, Op. cit., 1028

37

However built-in delays are quite difficult to implement as the delay needs to strike a fine

“balance between dampening the advantage of structural insiders—and still preserving the

benefits of HFT. To the extent that HFT is viewed positively as a boon for liquidity and a curb

on transaction costs, policy (and investors) might wish to maintain HFT’s presence in

markets.”169

Those delays correspond perfectly to the type of market response based on the velocity

principle. Here the emphasis is put on the speed of submission orders, which effectively prevent

structural insider dealing since HF traders are unable to trade on non-fully-public information

and would-be manipulators loses their ability to front-run orders and implement many

manipulative strategies.

Globally, built-in delays appear to be an appropriate response to HFT and an effective mean to

ensure a high degree of market integrity. However, the implications of built-in delays can only

be limited to the specific venues that have decided to implement that kind of structure. And that

sort of structure is not easily portable, as using delays to correct the structural imbalances is

both difficult and imprecise.170

A market-wide response would be more appropriate to tackle market abuse and ensure market

integrity. Especially if we consider the fragmentation of the modern marketplace. Requiring

every venue to set up delays would be near impossible. It however remains a very legitimate

and effective rethinking reorganisation of a market’s structure, if we consider the limited

impact.

Sub-section 2 – Frequent batch auctions

In a recent study, conducted by researchers of the University of Chicago and the University of

Maryland proposed a new market model based on “discrete-time trading” as an alternative to

the more common “continuous-time trading”. 171

As its name suggests, continuous-time trading refers to ability for computers to “buy or sell

stocks or other exchange-traded financial instruments at any instant during the trading day.”172

And according to the study, that model is the cause of the existing HFT arms race for speed.173

Indeed, in a marketplace where a gain in speed of a millisecond is potentially worth millions, it

169 Y. YADAV, Ibidem, p. 1029. 170 Y. YADAV, Ibidem, p. 1029. 171 E. BUDISH, P. CRAMTON and J. SHIM, Op. cit., pp. 1547 to 1621. 172 E. BUDISH, P. CRAMTON and J. SHIM, Ibidem, p. 1549. 173 E. BUDISH, P. CRAMTON and J. SHIM, Ibidem, p. 1549.

38

is natural to see investment firms and other market participants to jump in on the very lucrative

opportunity.

However, markets do not work exactly in continuous time, which, at HF creates mechanical

arbitrage opportunities. And “while there is an arms race in speed, the arms race does not

actually affect the size of the arbitrage prize; rather, it just continually raises the bar for how

fast one has to be to capture a piece of the prize.”174 This means that, according to their findings,

when performing arbitrage functions, HF traders, by increasing the speed at which they perform

this task, do not improve their ability to discover and eliminate inefficiencies – such as price

discrepancies – but merely improve their ability to do so ahead, and therefore at the exclusion,

of its competitors.

The alternative they offer is quite simple at first glance: the implementation of a “discrete-time

trading” model in which “the trading day is divided into extremely frequent but discrete time

intervals; to fix ideas, say, 100 milliseconds. All trade requests received during the same interval

are treated as having arrived at the same (discrete) time. Then, at the end of each interval, all

outstanding orders are processed in batch, using a uniform-price auction, as opposed to the

serial processing that occurs in the continuous market.”175 They call this new market design the

“frequent batch auctions”.

In practice, it would work like that: “at the end of each batch interval, the exchange batches all

outstanding orders—both new orders received during this interval, and orders outstanding from

previous intervals—and computes the aggregate demand and supply functions out of all bids

and asks, respectively. If demand and supply do not intersect, then there is no trade and all

orders remain outstanding for the next batch auction. If demand and supply do intersect, then

the market clears where supply equals demand, with all transactions occurring at the same

price—that is, at a “uniform price.””176

Just like the built-in delays, the frequent batch auctions help mitigate the disadvantages that

outsider investors face in contributing to price formation, as they will be better able to trade on

up-to-date information.177 Basically, the proposed market design somewhat amounts to the

introduction of system-wide delays model which eliminate the mechanical arbitrages and the

HFT arms race, which in turn enhances liquidity and, unless investors are extremely impatient,

174 E. BUDISH, P. CRAMTON and J. SHIM, Ibidem, p. 1552. 175 E. BUDISH, P. CRAMTON and J. SHIM, Ibidem, p. 1549. 176 E. BUDISH, P. CRAMTON and J. SHIM, Ibidem, p. 1595. 177 Y. YADAV, Op. cit., p. 1028.

39

improves social welfare. Discrete time makes tiny speed advantages orders of magnitude less

valuable, and the auction transforms competition on speed into competition on price.178

In a marketplace where the predominance of HF traders is ever-growing, frequent batch

auctions market design could practically supress the structural insider nature of HF traders as

orders are traded in batches instead of series. It is a structure that is worth looking into at EU

level.179 Especially since it differs from continuous-time trading model only in two designing

features: “time is treated as a discrete variable instead of a continuous variable. Second, orders

are processed in batch instead of serial—since multiple orders can arrive at the same (discrete)

time—using a standard uniform-price auction. All other design details are similar.”180 The fact

that this market design does not completely reshapes the market structure enables it not to

negate all aspects of HFT. Indeed, this design does not preclude HF traders from operating

market making or non-mechanical arbitrage strategies that are beneficial to the market and to

other investors. It is mainly targeted at manipulative and parasitic strategies.181

While this market structure is remains an economic model only viable in controlled conditions,

it goes to show that when the focus is put on speed instead of information, it is possible to

uncover responses to the issues raised HFT better suited to prevent market abuse than the new

framework adopted by the EU.

178 E. BUDISH, P. CRAMTON and J. SHIM, Ibidem, p. 1617. 179 H. BOONEN, Op. cit., available at www.nyujlb.com. 180 E. BUDISH, P. CRAMTON and J. SHIM, Op. cit., p. 1555. 181 C.R. KORSMO, Op. cit., p. 557 and f.

40

CONCLUSION

In this paper, we set out three different tasks, namely providing a general overview of the

incompatibility issues between HFT and market integrity, proposing a new paradigm through

the lens of which market integrity rules should be drafted in the context of HFT and assessing

ability of the market structure to ensure integrity in a predominantly led by HF traders.

First, we have highlighted time and again that HFT is a very specific kind of trading.

Characterised by a combination of co-location, direct feeds, and automated reactions to

incoming information, [through which] HF traders can enjoy first-sight and first-mover trading

advantages.”182 These exact features erode investor confidence in the market since they feel

like they cannot win against those privileged market participants.

To prevent such occurrences as the erosion of investor confidence, the EU has adopted MAR

in order to prevent abuses which are detrimental to market integrity. The European Lawmaker

attempted to restore confidence in the face of the rise of HFT activity in the markets by

prohibiting abusive behaviours such as insider dealing and market manipulation, which are

made easier by the sophisticated nature of HFT.

Second, we observed that MAR was supposed to fill the loopholes of MAD by making market

abuse rules applicable to HFT. However, the nature of HFT makes it very difficult to reconcile

HFT activity with market integrity from a conceptual point of view.

The irreconcilability of HFT and market integrity is highlighted by failure of MAR to actually

adapt the rules on insider dealing and market manipulation. On the one hand, the very nature of

HFT gives it structural advantages derived from the ability of HF traders to react to not-fully-

public information that harms market integrity the same way as does traditional insider dealing.

Yet, that so-called structural insider dealing is not addressed in MAR. Outsider investors are

left to fend for themselves, while HF traders get fat on profit returns. And on the other hand,

the sheer sophistication of HFT strategies and algorithm makes it extremely complex to

unambiguously detect. There as well, there were no real adaptation to the nature of HFT.

Based on the failure of MAR to take into account the specific features of HFT, we propose to

look at market integrity from a regulatory point of view through the lens of the “velocity

paradigm”. This paradigm is intended to offer a means to assess the adequacy of regulatory

182 Y. YADAV, Op. cit., p. 976.

41

measures on market integrity with the reality of the high-speed and high-frequency environment

in which HF traders evolve.

Third, we looked at market structures and how they could take into account the specific features

of HFT. We observed that if regulators are market participants (in the case of IEX) take into

account the specificities of HFT, we were able to observe measures and market design that are

able to effectively safeguard market integrity, without denying some of the benefits brought

about by HFT.

To conclude, we have observed throughout this paper that the new EU financial framework

which was for quite some time the talk of the town, while offering a serious and transparent

legal framework to regulate HFT in MiFID II, it also failed to offer any meaningful and

substantive advancement in the prevention of market abuse besides the shift from a Directive

that needs to be implemented to a directly applicable Regulation.

It feels like, when it comes t market integrity, the European Lawmaker is at a loss and only

manage to provide further transparency into the inner working of algorithms without

meaningful effects to the detection of market abuses, without which prevention is moot, as one

cannot prevent the adverse effects of a behaviour he cannot detect.

Since it is doubtful that the EU will endeavour to provide a new recast of MAR in the near

future, we cannot help but wonder if and how the EU will provide any means to ensure effective

market integrity because unless it is outright banned, HFT is here to stay. And it is likely that

HFT firms will keep taking advantage of the existing market structure to squeeze every possible

profit. And perhaps we will have to wait for HF traders to bring about the next big crisis in

order to see EU take effective and efficient steps to prevent disruptive behaviours.

42

BIBLIOGRAPHY

Legislations and official authorities’ documentation and decisions

EU legislation

- Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16

April 2014 on market abuse (market abuse regulation) and repealing Directive

2003/6/EC of the European Parliament and of the Council and Commission Directive.

- Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014

on markets in financial instruments and amending Directive 2002/92/EC and Directive

2011/61/EU.

- Directive 2004/39/CE of the European Parliament and of the Council of 21 April 2004

on markets in financial instruments amending Council Directives 85/611/EEC and

93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and

repealing Council Directive 93/22/EEC.

- Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003

on insider dealing and market manipulation (market abuse).

- Commission Delegated Regulation (EU) 2017/565 of 25 April 2016 supplementing

Directive 2014/65/EU of the European Parliament and of the Council as regards

organisational requirements and operating conditions for investment firms and defined

terms for the purposes of that Directive.

Italian legislation

- Law Decree of the Italian Minister of Economy and Finance of 21 February 2013

(Decreto del ministero dell'Economia e delle Finanze del 21 febbraio 2013).

ESMA

- ESMA Economic Report, “High frequency trading activity in EU equity markets”,

Number 1, 2014.

- ESMA Economic Report, “Order duplication and liquidity measurement in EU equity

markets”, Number 1, 2016.

US CTFC

- Bunge Global Markets, Inc., CFTC Docket No. 11-10 (March 22, 2011).

Scholarly and professional documentation

- ALDRIDGE (I.), High-Frequency Trading: A Practical Guide to Algorithmic Strategies

and Trading Systems, 2nd edition, Hoboken, N.J.: Wiley, 2013.

43

- AQUILINA (M.) and YSUSI (C.), “Are high-frequency traders anticipating the order

flow? Cross-venue evidence from the UK market”, FCA Occasional Papers, no 16,

2016.

- ARNUK (S.) and SALUZZI (J.), “Latency Arbitrage: The Real Power Behind Predatory

High Frequency Trading”, Themis Trading LLC White Papers, 4 December 2009.

- AITKEN (M.), CUMMING (D.) and ZHAN (F.), “Trade Size, High Frequency Trading,

and Co-Location Around the World”, European Journal of Finance, 2014, available at

www.papers.ssrn.com.

- AUSTEN (J.), “What Exactly Is Market Integrity: An Analysis of One of the Core

Objectives of Securities Regulation”, Wm. & Mary Bus. L. Rev., Vol. 8, 2017, pp. 215

to 240.

- BOONEN (H.), “High Frequency Trading, Electronic Frontrunning and Structural

Insider Trading Under the EU Market Abuse Regulation: Need for Reform?”, NYU J.

of Law & Bus., 27 November 2017, available at www.nyujlb.org.

- BRANDEIS (L.), Other people’s money, and how the bankers use it, New York: Stokes,

1914.

- BREALEY (R.A.) and MYERS (S.C.), Principles of Corporate Finance, New York:

McGraw-Hill, 1991.

- BROGAARD (J.A.), “High Frequency Trading and Its Impact on Market Quality”,

Northwestern University. Kellogg School of Management. Northwestern University

School of Law, 2010, available at www.fia.org.

- BUDISH (E.), CRAMTON (P.) and SHIM (J.), “The High-Frequency Trading Arms

Race: Frequent Batch Auctions as a Market Design Response”, Q. J. Econ., Vol. 130,

2015, pp. 1547 to 1621.

- BUSCH (D.), “MiFID II: regulating high frequency trading, other forms of algorithmic

trading and direct electronic market access”, L.F.M.R., Vol. 10, 2016, pp. 72 to 82.

- CARLTON (D.W.) and FISCHEL, “(D.R.), “The Regulation of Insider Trading”, Stan.

L. Rev., Vol. 35, 1982, pp. 857 to 895.

- CHABOUD (A.) et al., “Rise of the Machines: Algorithmic Trading in the Foreign

Exchange Market”, J. Finance, 2014, pp. 2045 to 2084.

- CLINCH (M.), “Italy launches tax on high-frequency transactions”, CNBC, 2

September 2013, available at www.cnbc.com.

- COLAERT (V.), “Recent trends in investor protection”, in Évolutions récentes en droit

financier, under the coordination of X. Dieux, Brussel: Bruylant, 2015.

44

- ČUK (T.) and VAN WAEYENBERGE (A.), “European Legal Framework for

Algorithmic and High Frequency Trading (Mifid 2 and MAR) A Global Approach to

Managing the Risks of the Modern Trading Paradigm”, Eur. J. Risk Regul., Vol. 9, 2018,

pp. 146 to 153.

- DE CORDT (Y.) and SCHAEKEN WILLEMAERS (G.), “Le concept de transparence

en droit financier”, Rev. prat. soc., 2006, p. 115.

- DIEUX (X.), “Du « marché efficient » au « marché fiable » ? À propos de la nouvelle

législation belge en matière de transparence” in Droit, morale et marché, Brussels:

Bruylant, 2013, pp. 439 to 454.

- EGGINTON (J.), VAN NESS (B.F.) and VAN NESS (R.A.), “Quote Stuffing”, 22

March 2016, available at www.papers.ssrn.com.

- FRIEDERICH (S.) and PAYNE (R.), “Order-to-trade ratios and market liquidity”,

Journal of Banking and Finance, 2015, pp. 214 to 223.

- GAI (J.), YAO (C.) and YE (M.), “The Externalities of High Frequency Trading”,

Working Paper, 15 March 2012, available at www.papers.ssrn.com

- HALDANE (A.), “Patience and Finance”, BIS Review, 114/2010.

- HASBROUCK, “The Best Bid and Offer: A Short Note on Programs and Practices”,

from the Finance Department, Stern School of Business, 14 October 2010, available at

www.papers.ssrn.com.

- HENDERSHOTT (T.) and RYORDAN (R.), “Algorithmic Trading and Information”,

Working Papers from NET Institute, No 09-08, 2009, available at www.netinst.org.

- KORSMO (C.R.), “High-Frequency Trading: A Regulatory Strategy”, U. Rich. L. Rev.,

Vol. 48, 2013, pp. 523 to 609.

- KRUDY (E.), “Is the May 6 "flash crash" in US stock markets the new normal?”,

Reuters, 10 June 2010, available at www.reuters.com.

- LEFÈVRE et al., “Les nouvelles règles en matière d’abus de marché”, R.D.C.-T.B.H.,

Vol. 5, 2017, p. 494.

- LERCH (M.P.), “Trading Activities”, in European Captial Markets Law, edited by R.

Veil, Oxford: Hart Publishing, 2nd Ed., 2017, pp. 459 to 522.

- MCNAMARA (S.), “The Law and Ethics of High-Frequency Trading”, Minn. J.L. Sci.

& Tech., Vol. 71, pp. 72 to 152.

- MENKVELD (A.), quoted in “Manipulatie in S&P-futures voorspelt weinig goeds voor

andere markten”, Financieele Dagblad, April 2015, available at www.fd.nl.

45

- MILLER (R.S.) and SHORTER (G.), “High-Frequency Trading: Background,

Concerns, and Regulatory Developments”, CRS Report, 19 June 2014.

- ONOFREI (A.), La négociation des instruments financiers : Au regard de la directive

MIF, Brussels: Larcier, 2012.

- PRICE (M.), “European banks pay $46.6 million to settle U.S. 'spoofing' charges”,

Reuters, 29 January 2018, available at www.reuters.com.

- SCOPINO (G.), “The (Questionable) Legality of High-Speed “Pinging” and “Front

Running” in the Futures Markets”, Conn. L. Rev., Vol. 47, 2015, pp. 607 to 697.

- STONE (G.), “SIP vs. Direct Feeds Latency – What are the Rules?”, Bloomberg White

Paper, 15 May 2014, available at www.bloomberg.com.

- WHEATLY, “Regulating high frequency trading”, speech made at the Global Exchange

and Brokerage Conference, New York, 4 June 2014, available at www.fca.org.uk.

- WYMEERSCH (E.), “Systemic Risk in the Securities Activities”, FLI UGent S&C,

2012-02, available at www.law.ugent.be.

- Y. YADAV, “Insider Trading and Market Structure”, UCLA L. Rev., 2016

- YADAV, “The Failure of Liability in Modern Markets”, Va. L. Rev., 2016, pp. 1031 to

1100.

- YADAV (Y.), “How Algorithmic Trading Undermines Efficiency in Capital Markets”,

Vand. L. Rev., 2015, pp. 1607 to 1671, pp. 968 to 1033.

- Clifford Chance, “The Italian Financial Transaction Tax Factsheet”, briefing note,

January 2013.

- Equity Markets Association, “The Importance of Market Data to the U.S. Equity

Markets”, EMA Whitepaper, 12 September 2012.

- FIA PTG, Debunking the Myths of High Frequency Trading, 2014, available at

www.fia.org.

46

TABLE OF CONTENT

INTRODUCTION ...................................................................................................................... 2

TITLE I – GENERAL CONCEPTS AND PRINCIPLES ......................................................... 4

CHAPTER I – HIGH-FREQUENCY TRADING ................................................................. 4

Section 1 – Technology and financial markets: a brief history of the rise of algorithmic

trading ................................................................................................................................. 4

Section 2 – High-frequency trading: automation at its paroxysm ...................................... 5

Sub-section 1 – Identifying high-frequency trading ...................................................... 6

Sub-section 2 – High-frequency trading’s strategies ..................................................... 8

Sub-section 3 – High-frequency trading in practice ..................................................... 10

CHAPTER II – MARKET INTEGRITY: THE STRONGHOLD AGAINST DISRUPTIVE

HIGH-FREQUENCY TRADING PRACTICES? ............................................................... 12

Section 1 – Market integrity ............................................................................................. 12

Section 2 – Market abuse ................................................................................................. 13

Sub-section 1 – Insider dealing .................................................................................... 13

Sub-section 2 – Market manipulation .......................................................................... 14

§1 – Quote stuffing ................................................................................................... 16

§2 - Pinging .............................................................................................................. 16

§3 - Spoofing ............................................................................................................ 17

§4 - Pump and dump ................................................................................................ 17

§5 – Accepted market practices ............................................................................... 17

TITLE II – FROM INFORMATION TO VELOCITY: A SHIFT IN FINANCIAL MARKETS’

PARADIGM ............................................................................................................................. 19

CHAPTER I – MiFID II, MARKET INTREGRITY AND HIGH-FREQUENCY TRADING:

TRANSPARENCY AT THE RESCUE? ............................................................................. 20

CHAPTER II – MARKET ABUSE RULES: A SWING AND A MISS? ........................... 23

Section 1 – Structural insider dealing .............................................................................. 23

Sub-section 1 – General concept .................................................................................. 23

47

Sub-section 2 – Structural insider dealing under the EU financial framework ............ 25

Section 2 – Detecting manipulative high-frequency trading strategies ............................ 26

CHAPTER III – MARKET INTEGRITY THROUGH THE LENS OF THE VELOCITY

PARADIGM ......................................................................................................................... 28

TITLE III – MARKET DESIGN IN MOVEMENT: THE ADEQUATE RESPONSE TO

HIGH-FREQUENCY TRADING? .......................................................................................... 31

CHAPTER I – RESPONSES TO HIGH-FREQUENCY TRADING UNDER THE

CURRENT MARKET STRUCTURE ................................................................................. 31

Section 1 – Ban on high-frequency trading ...................................................................... 32

Section 2 – Taxation of high-frequency trading activity: the Italian case ....................... 32

CHAPTER II – RETHINKING MARKET STRUCTURE ................................................. 34

Section 1 – Recognition of structural insider dealing: taking a stance ............................ 34

Section 2 – Restoration of the level-playing field ............................................................ 35

Sub-section 1 – Equalising access: the case of IEX ..................................................... 36

Sub-section 2 – Frequent batch auctions ...................................................................... 37

CONCLUSION ........................................................................................................................ 40

BIBLIOGRAPHY .................................................................................................................... 42

Legislations and official authorities’ documentation and decisions .................................... 42

EU legislation ............................................................................................................... 42

Italian legislation .......................................................................................................... 42

ESMA ........................................................................................................................... 42

US CTFC ...................................................................................................................... 42

Scholarly and professional documentation .......................................................................... 42

TABLE OF CONTENT ........................................................................................................... 46

48