Civil Law and Algorithmic Trading Market Manipulation Claims in Europe

1. Introduction

Algorithmic trading market manipulation claims arise when automated trading systems, high-frequency trading algorithms, or AI-driven trading strategies are alleged to have distorted market prices, trading volumes, liquidity, or the appearance of supply and demand, causing losses to other market participants.

European law expressly recognises that market manipulation can occur through electronic and algorithmic trading. ESMA has specifically identified practices such as spoofing, layering and quote stuffing as risks associated with algorithmic trading. (ESMA)

Under MiFID II, algorithmic trading includes situations where a computer algorithm automatically determines matters such as whether to initiate an order, its timing, price, quantity, or how to manage the order after submission, with limited or no human intervention. (ESMA)

However, an important legal distinction must be made:

Market-abuse rules establish prohibited conduct and regulatory sanctions, while a private investor's claim for compensation generally depends on the applicable national civil-law regime.

Thus, a finding of market manipulation does not automatically produce an EU-wide private damages award.

2. Meaning of Algorithmic Trading Market Manipulation

Algorithmic market manipulation occurs when an automated trading strategy is used to create a false or misleading appearance concerning:

supply;

demand;

trading activity;

price;

liquidity;

market depth;

volatility.

The algorithm may execute hundreds or thousands of orders in fractions of a second.

Typical techniques include:

Spoofing

Layering

Quote stuffing

Momentum ignition

Marking the close

Wash trading

Ramping

Cross-market manipulation

Manipulative order cancellation

Artificial liquidity creation

3. European Regulatory Framework

The central European framework is the Market Abuse Regulation (MAR), Regulation (EU) No. 596/2014.

MAR is designed to protect:

market integrity;

investor confidence;

fair price formation.

ESMA explains that market abuse includes insider dealing, unlawful disclosure of inside information and market manipulation. (ESMA)

Importantly, MAR expressly contemplates manipulation through electronic means.

ESMA has stated that Article 12(2)(c) MAR covers market manipulation connected with electronic means of trading, including algorithms and high-frequency trading strategies. (ESMA)

4. Algorithmic Trading Under MiFID II

MiFID II imposes specific obligations on firms engaged in algorithmic trading.

An investment firm using algorithmic trading must have:

resilient systems;

sufficient capacity;

trading thresholds;

risk controls;

controls against erroneous orders;

controls preventing disorderly markets;

controls preventing use of systems contrary to MAR;

business-continuity arrangements;

appropriate testing and monitoring. (ESMA)

Therefore, algorithmic market manipulation can involve two different legal problems:

A. Manipulative conduct

The algorithm deliberately or knowingly creates a false or misleading market.

B. Algorithmic-control failure

The firm fails to maintain adequate controls and its algorithm creates or contributes to disorderly trading.

These should not automatically be treated as the same type of liability.

5. Civil Claim Structure

A private investor seeking compensation may need to establish:

1. Unlawful conduct

Example:

Algorithm deliberately creates artificial demand.

2. Breach of a legal duty

For example:

MAR;

national securities law;

tort/delict law;

contractual obligations.

3. Causation

The claimant must connect the manipulation with the loss.

4. Damage

Examples:

inflated purchase price;

depressed sale price;

trading losses;

lost investment value;

transaction costs.

5. Recoverability

The relevant national law determines which losses can be recovered.

6. Case Law 1 — Spector Photo Group, C-45/08

Court: CJEU
Date: 23 December 2009

Spector Photo Group NV and Chris Van Raemdonck v CBFA concerned European market-abuse law and insider dealing.

The CJEU interpreted the Market Abuse Directive and addressed the use of inside information and regulatory sanctions. (EUR-Lex)

Although it was not an algorithmic market-manipulation damages case, it is an important foundation for understanding the EU market-abuse framework.

Relevance to algorithms

Algorithmic systems can execute trades automatically once particular conditions are satisfied.

The legal question remains whether the conduct falls within prohibited market-abuse rules; automation does not itself make conduct lawful.

Principle

Automated execution does not remove the application of market-abuse law.

7. Case Law 2 — Geltl v Daimler, C-19/11

Court: CJEU
Date: 28 June 2012

The case concerned the meaning of “precise information” in the market-abuse framework.

The CJEU held that intermediate steps in a prolonged process can themselves constitute inside information when the relevant legal conditions are satisfied. (EUR-Lex)

Relevance to algorithmic manipulation

Algorithms process information at extremely high speed.

An automated trading strategy may react to:

announcements;

order-book information;

corporate developments;

market signals;

intermediate events.

The case therefore helps explain why courts must examine the precise informational circumstances surrounding algorithmic trading.

Limitation

Geltl is an insider-dealing case, not a direct algorithmic-manipulation damages case.

It is best treated as related market-abuse jurisprudence.

8. Case Law 3 — Lafonta v AMF, C-628/13

Court: CJEU
Date: 11 March 2015

The CJEU considered the meaning of information of a precise nature under the market-abuse framework.

It held that information does not have to permit a determination that its price effect will necessarily be in a particular direction in order to satisfy the relevant definition of precise information. (Infocuria)

Algorithmic significance

Trading algorithms may operate on information that does not produce a certain directional result.

For example:

Algorithm receives information suggesting an increased probability of volatility.

It may immediately alter:

order quantity;

execution speed;

price;

market exposure.

Lafonta demonstrates why market-abuse analysis cannot simply depend upon whether the information guaranteed a particular price movement.

9. Case Law 4 — VD and SR, Joined Cases C-339/20 and C-397/20

Court: CJEU, Grand Chamber
Date: 20 September 2022

These cases concerned market-abuse investigations and the ability of financial authorities to obtain traffic and communications data.

The CJEU examined the tension between:

financial-market integrity;

investigative powers;

privacy;

protection of communications;

personal data.

It held that general and indiscriminate retention of traffic data could not be justified simply by relying on market-abuse investigation objectives. (EUR-Lex)

Importance for algorithmic manipulation claims

Algorithmic manipulation investigations often depend upon:

trading records;

communications;

system logs;

order messages;

timestamps;

electronic communications.

The case demonstrates that obtaining such evidence must still respect EU privacy and data-protection requirements.

Principle

Effective market-abuse enforcement does not create unlimited surveillance powers.

10. Case Law 5 — DB v Consob, C-481/19

Court: CJEU, Grand Chamber
Date: 2 February 2021

This case concerned the right to remain silent and protection against self-incrimination in market-abuse proceedings.

The CJEU considered administrative sanctions of a criminal nature and Articles 47 and 48 of the EU Charter. (EUR-Lex)

Algorithmic relevance

Suppose regulators investigate:

algorithmic spoofing.

They may seek:

developer communications;

trading instructions;

strategy documentation;

explanations of algorithm parameters;

communications between traders and programmers.

The case demonstrates that regulatory investigations must respect fundamental procedural rights.

Civil significance

Evidence obtained in regulatory proceedings may become relevant to later civil litigation, but its use and admissibility depend upon the applicable procedural law.

11. Case Law 6 — Di Puma and Zecca, Joined Cases C-596/16 and C-597/16

Court: CJEU, Grand Chamber
Date: 20 March 2018

These cases concerned the interaction between:

criminal proceedings;

administrative market-abuse penalties;

the ne bis in idem principle.

The Court held that EU law can prevent subsequent administrative proceedings of a criminal nature after a final criminal acquittal establishing that the relevant acts were not proven. (EUR-Lex)

Relevance to algorithmic manipulation

An algorithmic manipulation incident can potentially produce:

regulatory proceedings;

criminal proceedings;

civil claims.

These proceedings cannot simply be treated as completely independent where fundamental-rights restrictions apply.

Principle

Multiple enforcement mechanisms must be coordinated with fundamental procedural guarantees.

12. Case Law 7 — Sigríður Elín Sigfúsdóttir v Iceland

Court: ECtHR

This market-manipulation-related case concerned procedural fairness in proceedings arising from financial-market conduct.

It is useful for understanding that market-abuse enforcement must comply with Convention fair-trial guarantees.

Algorithmic relevance

An individual or company accused of algorithmic manipulation must be able to challenge:

technical evidence;

expert evidence;

trading data;

market analysis;

interpretation of algorithmic conduct.

This is especially important because algorithmic trading evidence can be extremely complex.

13. Case Law 8 — Nodet v France

Court: ECtHR
Application: No. 47342/14
Date: 6 June 2019

The case concerned overlapping proceedings relating to market manipulation.

The ECtHR found a violation of the ne bis in idem principle because the administrative and criminal proceedings did not have a sufficiently close connection in substance and time.

Relevance to algorithmic trading

One algorithmic trading incident might generate:

regulator investigation → administrative penalty → criminal proceedings → investor litigation.

Nodet demonstrates that the relationship between different proceedings can itself raise Convention issues.

Important limitation

Nodet does not establish a private damages rule for algorithmic trading.

It is a procedural market-abuse authority.

14. Case Law 9 — Spector Photo Group and Market-Abuse Enforcement

The significance of Spector extends beyond the individual facts because the CJEU explained the EU market-abuse framework as a mechanism designed to protect:

financial-market integrity;

investor confidence.

The judgment confirms that EU market-abuse legislation permits effective sanctions while requiring compliance with defence rights. (curia)

For algorithmic trading, this means that an automated strategy cannot be assessed solely by looking at whether individual orders were technically valid.

The broader trading pattern can matter.

15. Case Law 10 — 2026 Brännelius, C-229/24

Court: CJEU
Date: 16 April 2026

This recent case concerned market abuse under MAR and the concept of inside information.

The CJEU considered circumstances involving a public procurement decision, early sale of shares and the meaning of inside information under Article 7 MAR. (Infocuria)

Relevance to algorithmic systems

Algorithms can automatically react to information before it becomes widely known.

The case illustrates the continuing importance of identifying:

what information existed;

when it existed;

whether it was public;

whether it was sufficiently precise;

how the information could affect trading.

Although Brännelius is an insider-dealing case rather than a direct algorithmic manipulation case, it is relevant to the information environment in which automated strategies operate.

16. Algorithmic Spoofing

Spoofing occurs when a trader places orders without genuine intention of executing them, intending to create a misleading impression of supply or demand.

Example:

Algorithm places 10,000 large buy orders.

Other market participants see strong apparent demand.

The price rises.

The algorithm then cancels the large orders and sells at the higher price.

Potential civil claimant:

Investor bought at the artificially elevated price and later suffered a loss.

Possible legal questions:

Was the order genuinely intended to execute?

Was there a manipulative purpose?

Did the conduct affect the market?

Did the claimant rely on the distorted market?

What was the counterfactual price?

ESMA has expressly identified spoofing and layering among the manipulation risks associated with algorithmic trading. (ESMA)

17. Algorithmic Layering

Layering is related to spoofing but involves multiple levels of orders.

Example:

Sell side €101 — genuine order €102 — genuine order Buy side €99 — large artificial order €98 — large artificial order €97 — large artificial order

The artificial orders create the appearance of substantial demand.

The algorithm then executes genuine trades at advantageous prices.

18. Quote Stuffing

Quote stuffing involves sending an extremely large number of orders or modifications, often with rapid cancellation.

Potential purposes can include:

creating congestion;

slowing competitors;

creating misleading market activity;

obscuring genuine trading signals.

Not every high message rate is manipulation.

The legal question depends upon the facts and applicable MAR indicators.

19. Momentum Ignition

A trader's algorithm may execute transactions designed to trigger other algorithms.

For example:

Algorithm A

↓

creates artificial buying pressure

↓

Algorithms B, C and D detect momentum

↓

automatically buy

↓

price rises

↓

Algorithm A sells.

This creates a difficult causation problem.

The claimant must establish that the manipulative activity materially affected the price at which the claimant traded.

20. Marking the Close

An algorithm may place trades close to the end of a trading session to influence:

closing price;

benchmark;

portfolio valuation;

derivatives settlement.

A small price movement at the close can affect large portfolios.

Potential damages therefore may substantially exceed the value of the individual manipulative transaction.

21. Wash Trading

Wash trading involves transactions that create artificial trading activity without genuine transfer of economic risk.

For example:

Algorithm A sells to an account controlled by the same beneficial owner.

The transaction appears in market data as trading activity but may not represent genuine market demand.

Potential consequences include:

artificial volume;

distorted price discovery;

misleading investors.

22. Algorithmic Cross-Market Manipulation

An algorithm may manipulate one market to influence another.

Example:

Manipulate futures price → influence spot market → profit from options.

This creates complicated causation questions because the claimant's loss may occur in a different market from the manipulative conduct.

23. Civil Liability vs Regulatory Liability

This distinction is extremely important.

Regulatory proceeding

Regulator asks:

Did the participant violate MAR or financial-market rules?

Possible consequences:

administrative fine;

trading restrictions;

supervisory action;

criminal prosecution where applicable.

Civil proceeding

Investor asks:

Did the unlawful conduct cause me a recoverable loss?

Possible remedies:

damages;

restitution;

rescission where available;

other national-law remedies.

A regulatory penalty is not automatically compensation to investors.

24. Is There an EU-Wide Civil Action for Market Manipulation?

Generally, no single EU-wide private damages action applies to every market-manipulation loss.

MAR harmonises market-abuse prohibitions and enforcement, but the private-law consequences are substantially shaped by:

national tort law;

contract law;

securities law;

procedural rules;

causation rules;

limitation periods;

collective-redress mechanisms.

Therefore:

MAR violation and private civil liability must be analysed separately.

25. Elements of a Private Claim

A claimant may need to establish:

A. Manipulative conduct

For example:

spoofing or layering.

B. Unlawfulness

The conduct falls within the relevant prohibition.

C. Loss

For example:

shares purchased at an artificially inflated price.

D. Causation

The manipulation materially contributed to the transaction price.

E. Recoverability

National law recognises the claimed type of damage.

26. The Counterfactual Price

This is often the central economic question.

The claimant purchased at:

€120

The alleged manipulation caused the price to rise.

The claimant argues that the genuine market price would have been:

€100

Potential price distortion:

€20

The claimant might therefore argue that the manipulation caused a €20 per-share loss.

But the defendant may argue:

The €20 difference was caused by legitimate market developments.

Expert economic evidence becomes crucial.

27. Causation in Algorithmic Manipulation

Causation can be especially complicated because markets are influenced by many variables:

macroeconomic news;

company announcements;

interest rates;

other algorithms;

institutional trading;

market sentiment;

liquidity;

geopolitical events.

Therefore:

Price movement alone does not prove manipulation.

A claimant must connect the unlawful conduct to the particular loss under the applicable national law.

28. Event Study Evidence

Experts may use an event study to examine price movements.

Simplified:

Expected price

versus

Actual price

around the suspected manipulation period.

The expert may investigate:

abnormal returns;

trading volume;

order-book changes;

volatility;

timing;

cancellation patterns.

This can help establish whether the suspected conduct materially affected the market.

29. Order-Book Evidence

Algorithmic manipulation litigation can require analysis of:

order submission;

order cancellation;

order modification;

execution;

timing;

order size;

order position;

price levels.

This creates a highly technical evidentiary environment.

30. Algorithm Logs

Important evidence may include:

source code;

trading rules;

configuration files;

model versions;

order logs;

timestamps;

kill-switch records;

risk limits;

communications;

human overrides.

The absence of proper records can itself become important to the evidentiary dispute, depending on the applicable national rules.

31. Algorithmic Error vs Manipulation

Not every abnormal algorithmic trading event is manipulation.

Consider:

Trading algorithm accidentally sends 50,000 orders because of a software bug.

Possible issues:

operational negligence;

failure of risk controls;

erroneous orders;

disorderly market.

But this is not necessarily the same as deliberate manipulation.

MiFID II specifically requires algorithmic trading firms to maintain controls designed to prevent erroneous orders and systems that could contribute to disorderly markets. (ESMA)

Thus, two distinct civil theories may exist:

Intentional manipulation

Algorithm deliberately designed or used to distort the market.

Algorithmic negligence

Firm failed to control a defective algorithm.

32. Human Responsibility for Automated Trading

An algorithm cannot normally be treated as an independent legal actor.

Responsibility may potentially attach to:

investment firm;

trader;

algorithm developer;

portfolio manager;

compliance officer;

trading venue;

direct-electronic-access provider.

The exact allocation depends on the applicable legal duties.

MiFID II requires investment firms engaging in algorithmic trading to maintain systems and controls preventing conduct contrary to MAR. (ESMA)

33. Direct Electronic Access

Direct electronic access creates additional risk because clients can send orders directly to a venue through another firm's infrastructure.

MiFID II requires appropriate controls concerning:

client suitability;

trading thresholds;

credit thresholds;

monitoring;

prevention of disorderly trading;

compliance with MAR. (ESMA)

This can become important when determining whether the intermediary or client bears responsibility.

34. Trading Venue Responsibility

Trading venues also have obligations concerning algorithmic trading.

European rules require systems capable of dealing with:

disorderly markets;

algorithm testing;

order-to-trade ratios;

order-flow management;

system capacity.

(ESMA)

Therefore, civil claims may potentially raise questions concerning the conduct of a venue as well as the trading firm.

35. Algorithmic Trading and Market Liquidity

Algorithms often provide substantial liquidity.

Consequently, courts should distinguish:

legitimate high-frequency market making

from

manipulative artificial liquidity.

A large number of orders does not by itself establish manipulation.

The surrounding facts matter:

intention;

order execution;

cancellation pattern;

price effect;

market conditions;

legitimate strategy;

accepted market practices.

ESMA notes that accepted market practices can provide a defence to allegations of manipulation where the dealings were carried out for legitimate reasons and in accordance with the relevant practice. (ESMA)

36. Intent and Algorithmic Manipulation

One difficult question is:

Whose intention matters?

Possible answers can involve:

programmer;

trader;

investment firm;

portfolio manager.

An algorithm may independently execute thousands of transactions.

But the law generally examines the conduct and responsibility of the relevant natural or legal persons rather than treating the algorithm itself as having legal intent.

37. Black-Box Algorithms

A claimant may not know why an algorithm behaved in a particular manner.

This creates an information imbalance.

The defendant may possess:

code;

proprietary models;

logs;

strategy documentation;

internal communications.

The claimant may have only:

“The price suddenly moved.”

Therefore, disclosure and expert evidence become central.

38. Privacy Limits on Market-Abuse Investigations

VD and SR is important because market-abuse authorities may need electronic communications data, but the CJEU stressed the privacy and proportionality constraints applicable to such investigative powers. (EUR-Lex)

For civil litigation, this creates a balancing problem:

Need for evidence

versus

Privacy and data protection.

39. Regulatory Evidence in Civil Litigation

Suppose a financial regulator concludes:

Algorithm X manipulated the market.

An investor later files a damages claim.

The regulatory finding may be highly relevant evidence, but it does not automatically answer every civil-law question.

The civil court may still have to determine:

claimant's loss;

causation;

counterfactual price;

recoverable damages;

limitation;

contributory conduct.

40. Multiple Victims

Algorithmic market manipulation can harm thousands of investors simultaneously.

This creates potential collective litigation.

For example:

Algorithm inflates share price for three days → thousands of investors purchase → price subsequently returns to normal.

Potential legal mechanisms include:

representative actions;

collective proceedings;

group litigation;

shareholder actions;

national class-action mechanisms.

The availability and structure differ across European jurisdictions.

41. Damages

Potential forms of loss include:

Direct trading loss

Difference between purchase/sale price and genuine market value.

Transaction costs

Brokerage and trading costs caused by the distorted market.

Loss of opportunity

Potentially recoverable only where recognised by national law and adequately proven.

Portfolio loss

Where manipulation affects multiple positions.

Consequential economic loss

Subject to national rules concerning remoteness and foreseeability.

42. Contributory Conduct

A defendant may argue that the claimant:

should not have relied upon the price;

failed to mitigate loss;

independently made a speculative decision;

had access to contrary information.

The legal relevance of such arguments depends upon national law.

43. Defences

Potential defences include:

1. No manipulation

The transactions were legitimate.

2. Accepted market practice

The conduct fell within an applicable accepted market practice.

3. No causal connection

The claimant's loss resulted from independent market factors.

4. No recoverable damage

The alleged loss is too remote or speculative.

5. Algorithmic malfunction without legal fault

The firm may argue that it maintained appropriate controls and acted reasonably.

6. Claimant's own conduct

The claimant's own decisions contributed to the loss.

44. Algorithmic Market Manipulation and Fundamental Rights

Market-abuse enforcement must respect:

fair-trial rights;

defence rights;

privilege against self-incrimination;

privacy;

data protection.

Consob, C-481/19, is particularly relevant to the right to silence. (EUR-Lex)

VD and SR is particularly relevant to privacy and communications data. (EUR-Lex)

Di Puma and Zecca and Nodet illustrate the importance of avoiding incompatible duplication of punitive proceedings. (EUR-Lex)

45. Important Distinction Between Six Types of Claims

One incident can generate several of these claims.

46. Consolidated Case-Law Table

47. Most Important Legal Distinction

There are three different questions in an algorithmic trading dispute:

Question 1 — Was there market manipulation?

This is principally a MAR/financial-regulatory question.

Question 2 — Was the trading firm legally responsible?

This involves regulatory, contractual, tort and potentially corporate-liability principles.

Question 3 — Did a particular investor suffer recoverable damage?

This is principally a private-law question.

The answer to Question 1 does not automatically answer Questions 2 and 3.

48. Hypothetical Example

Suppose Algorithm A operates on a European exchange.

It places:

20,000 artificial buy orders;

cancels them milliseconds later;

creates apparent demand;

causes the price to rise from €50 to €58;

sells genuine holdings at €58.

After the algorithm stops, the price returns to €50.

An investor bought 10,000 shares at €58.

Alleged loss

€8 × 10,000 = €80,000

But the investor must still establish, under the applicable national law:

that the conduct was unlawful;

that the market was artificially distorted;

that the algorithm caused the relevant price distortion;

that the claimant's purchase was affected;

that €80,000 represents legally recoverable damage.

49. Role of Expert Economists

Expert evidence may examine:

normal market price;

abnormal price movement;

trading volume;

order-book depth;

algorithmic activity;

comparable securities;

market-wide movements;

timing;

liquidity;

volatility.

The central question becomes:

What would the market price probably have been absent the alleged manipulation?

That is usually more complicated than simply comparing the purchase price with the later market price.

50. Role of Technical Experts

Technical experts may examine:

source code;

algorithm architecture;

order-generation rules;

cancellation logic;

latency;

risk controls;

kill switches;

testing;

model updates;

logs.

This can determine whether the conduct was:

intentional;

negligent;

accidental;

caused by software malfunction;

caused by external interference.

51. Importance of MiFID II Risk Controls

MiFID II specifically requires algorithmic trading firms to have controls designed to prevent:

erroneous orders;

disorderly markets;

use of trading systems contrary to MAR.

(ESMA)

This is important for civil litigation because a claimant may argue:

The firm failed to implement safeguards that European financial-market law specifically expects algorithmic traders to maintain.

The precise civil-law consequence of that regulatory breach, however, depends on the national legal system.

52. Civil Liability for Algorithmic Trading Errors

Not every algorithmic loss involves manipulation.

Consider:

Algorithm accidentally buys €100 million of shares because of a coding error.

Possible legal issues include:

negligence;

inadequate risk controls;

contractual liability;

exchange rules;

market-disruption obligations.

This is different from:

Algorithm deliberately generates false demand to move the market.

The second is a market-manipulation problem.

53. European Civil-Law Model

The European framework can therefore be represented as:

Algorithm

↓

Trading activity

↓

MAR / MiFID II compliance

↓

Potential manipulation or control failure

↓

Market distortion

↓

Investor transaction

↓

Economic loss

↓

National civil-law claim

↓

Damages

This explains why market-abuse regulation and private civil liability must be analysed together but not conflated.

54. Key Legal Principles

Principle 1

Algorithms are expressly recognised as capable of facilitating market manipulation.

ESMA specifically identifies algorithmic and high-frequency trading as environments in which manipulation risks arise. (ESMA)

Principle 2

Not every algorithmic error is manipulation.

Intentional manipulation and negligent system failure are different legal problems.

Principle 3

High-frequency trading is not itself unlawful.

Legitimate algorithmic trading is permitted.

Principle 4

Spoofing and layering can constitute manipulation depending on the facts and applicable legal criteria.

Principle 5

Market manipulation does not automatically create an EU-wide damages action.

Principle 6

Private compensation normally requires proof of legally recoverable damage and causation under applicable national law.

Principle 7

Regulatory proceedings and civil claims serve different functions.

Principle 8

Electronic evidence is central to algorithmic manipulation litigation.

Principle 9

Privacy and fair-trial rights constrain regulatory investigations.

Principle 10

MiFID II imposes specific systems-and-controls obligations on algorithmic traders. (ESMA)

55. Short Exam Answer

Algorithmic trading market manipulation claims in Europe arise where automated or high-frequency trading systems are used to create false or misleading conditions concerning price, supply, demand, liquidity or trading activity. European law expressly recognises algorithmic manipulation. MAR prohibits market manipulation, while MiFID II requires algorithmic trading firms to maintain effective systems and controls preventing erroneous orders, disorderly markets and conduct contrary to MAR. (ESMA)

Important authorities include Spector Photo Group, C-45/08, concerning the EU market-abuse framework; Geltl, C-19/11, concerning precise information and intermediate steps; Lafonta, C-628/13, concerning the meaning of precise information; VD and SR, C-339/20 and C-397/20, concerning market-abuse investigations and communications data; DB v Consob, C-481/19, concerning the right against self-incrimination; Di Puma and Zecca, C-596/16 and C-597/16, concerning ne bis in idem; Nodet v France, concerning duplicate market-abuse proceedings; and Brännelius, C-229/24, concerning inside information under MAR. (EUR-Lex)

For a private investor, however, proving regulatory manipulation is only part of the case. The claimant generally must also establish causation, actual recoverable loss and the availability of a private-law remedy under the relevant national legal system.

56. Ultra-Short Revision

Algorithmic Trading Manipulation

Algorithm → Artificial orders → Market distortion → Investor transaction → Loss

Remember:

MAR → market manipulation

MiFID II → algorithmic-trading controls

Spoofing → false orders

Layering → multiple artificial order levels

Quote stuffing → excessive order activity

Marking the close → influencing closing price

Spector → market-abuse framework

Geltl → precise information

Lafonta → price-direction certainty not required

VD & SR → communications/data evidence

Consob → right against self-incrimination

Di Puma & Zecca → ne bis in idem

Nodet → duplicate proceedings

Brännelius → MAR information analysis

Core principle:

European law treats algorithmic trading as a legitimate financial technology but subjects it to market-integrity and systems-control requirements; when automated trading crosses into manipulation, private compensation still requires a separate analysis of national civil liability, causation and investor loss.