Financial market manipulation with particular emphasis on algorithmic trading

The technological progress we have seen in recent years has led to the creation of increasingly effective tools to facilitate functioning in the modern world.

Constantly increasing computing power supports the development of artificial intelligence, which, using machine learning, based on so-called learning algorithms, is able to provide, from a human point of view, fully rational and beneficial solutions for everyday application.

According to Deloitte's estimates, 2017, In 2021 This appropriation is intended to cover the following expenditure: 57,000,000,000 USD. The use of modern technologies also gives rise to the risk of fraud, including through financial market manipulation.

1. Financial market manipulation

The manipulation of the financial market, in addition to insider trading, should be considered to be the most dangerous category of fraud which leads to financial losses for investors and thus also to loss of confidence in the market.[1].

According to the dictionary, the term “manipulation” means the use of circumstances, bending or twisting facts to prove their right or influencing others’ views and behaviour.[2].

In literature, the concept of manipulation of financial instruments has not yet been defined uniformly, as it is difficult to define them in such a way that they are unambiguously and exhaustive describe all types of behaviour of this type on the market[3].

However, we can agree that manipulating the course of a financial instrument involves using various unfair measures to influence its price for profit[4].

The literature takes the division into transactional and information manipulations. Transactional manipulation shall consist in making commercial decisions that mislead other market participants as to the demand, supply or price of the financial instrument concerned.

Information manipulation, on the other hand, is when investors are misled about demand, supply or price as a result of spreading false information. Information manipulation often leads to an unnatural or artificial level of price maintenance for a financial instrument.

Untrue information may concern both the financial instrument and the issuer[5].

Regulation (EU) 596/2014 to 16 April 2014 on market abuse (hereinafter referred to as the MAR Regulation) regulates which actions can be considered market manipulation and which should be considered such manipulation. This Regulation lists two activities which may be considered to be transaction manipulation; and two, which may be considered information manipulation[6].

first two transaction manipulation activities, regulated under Article 12(1) point (a) and b MAR Regulations concern:

(a) the conclusion of transactions, the submission of orders or other behaviours which:

1. Give or could give false or misleading signals as to the supply or demand of a financial instrument, a related commodity contract on the spot market or an auctioned product based on emission allowances or the price thereof; or
2. Keep or keep price one or a number of financial instruments, a related commodity contract on the spot market or an auctioned product based on emission allowances at an unnatural or artificial level; unless the person making the transaction, placing the order for a transaction or taking any other conduct proves that the transaction, order or conduct has occurred for reasonable reasons and is in line with the accepted market practices established according to the Article 13 MAR Regulations;

(b) the conclusion of transactions, the submission of orders or other activities or behaviours affecting or likely to affect the price one or several financial instruments, a related commodity contract on the spot market or an auctioned product based on emission allowances, related to the use of fictitious tools or other forms of misleading or deceit.

The EU legislature provides that, in terms of manipulation, account should be taken of a catalogue of circumstances indicating the manipulation of the transactions or orders of investors, as well as of the behaviours listed above. In Annex I point (a) to the MAR Regulation and In Annex II section 1 of Regulation (EU) 2016/522 7 .

Information manipulation activities have been regulated under Article 12(1) point (c) i d MAR Regulations as follows:

(c) the dissemination through the media, including the internet, or other means, of information which gives or could give false or misleading signals as to the supply or demand of a financial instrument, the related commodity contract on the spot market or the auctioned product based on emission allowances, or as regards the price thereof, or ensure that the price persists or can ensure the maintenance of the price one or several financial instruments, a related commodity contract on the spot market or an auctioned product based on emission allowances at an unnatural or artificial level, including spreading rumors where the person distributing the information knew or should have known that the information was false or misleading;

(d) the transmission of false or misleading information, or the provision of false or misleading benchmark data, if the data exporter or the data provider knew or should have known that they were false or misleading, or any other behaviour that manipulated the calculation of the benchmark[8].

The MAR Regulation also provides for a catalogue of behaviours to be considered market manipulation. According to Article 12(2) MAR is considered to be one of the following:

(a) the conduct of a person or persons acting jointly, with a view to maintaining a dominant position in the supply or demand of a financial instrument, related commodity contracts on the spot market or auctioned products based on emission allowances which result in or may result in, directly or indirectly, a level of sale or purchase prices or create or may create unfair trading conditions;

(b) the acquisition or disposal of financial instruments at the opening or closing of a market which results in or may result in misleading investors with prices made public, including opening and closing prices;

(c) the submission of orders on a trading venue, including their cancellation or amendment, by any available means of trade, including electronic means,

such as algorithmic trading and high-frequency trading strategies, and which triggers one of the effects in question Under section 1 point (a) or (b) by:

• 1. Interference or delay in the operation of transactions on a trading venue or the likelihood of causing them;
• 2. Obstruction of the identification of real orders on a trading venue or the likelihood of hindering that identification, in particular by submitting orders that result in the execution or destabilisation of the order book; or
• 3. Creating or likely to create a false or misleading signal in terms of supply or demand for a financial instrument or its prices, in particular by making orders to initiate or worsen the trend;

(d) the use of occasional or regular access to traditional or electronic media to give an opinion on a financial instrument, a related commodity contract on the spot market or an auctioned product based on emission allowances (or indirectly on its issuer) after taking a position on a given financial instrument, a related commodity contract on the spot market or an auctioned product based on emission allowances, and then profiting from opinions on the price of that instrument, a related commodity contract on the spot market or an auctioned product based on emission allowances, without making public the existence of a conflict of interest in an appropriate and effective manner;

(e) the acquisition or disposal on the secondary market of emission allowances or related derivatives before an auction organised in accordance with Regulation (EU) 1031/2010, with the effect of establishing the settlement auction price of products auctioned at an unnatural or artificial level or of misleading auction bidders[9].

In the bill of 29 July 2005 on trading in financial instruments 10 (hereafter as u.o.i.f.), criminal liability is foreseen for making/existing market manipulation. According to Article 183 u.o.i.f.

a person who, contrary to the prohibition in question, under Article 15 The MAR Regulation shall manipulate the said under Article 12 MAR Regulation, subject to fine until 5,000,000 PLN or punishes imprisonment from 3 months to years 5, or both of these penalties together, while a person who enters into agreement with another person intended to manipulate shall be fined until 2,000,000 PLN.

2. Algorithmic trade

The term algorithmic trade (also called algo trading) is defined in both Polish and Community law. According to Article 3(2b) u.o.i.f.

by this term, the acquisition or disposal of financial instruments by means of a computer algorithm that automatically identifies the individual parameters of the acquisition or disposal orders, including the moment of the submission of the order, its validity, the price or number of instruments subject to the order or the manner in which the order is managed after its submission, without human participation or with limited human participation within the meaning of the contract, shall be understood to mean the purchase or disposal of financial instruments by means of a computer algorithm, Article 18 Regulation (EU) 2017/565 11 , subject to the fact that it is not an algorithmic trade, the use of automatic systems used solely for the purpose of redirecting orders between trading venues in financial instruments, the processing of orders not involving the identification of any transaction parameters, the confirmation of orders or the processing of post-trade transactions.

The algorithmic trade is done using high technology. Its advantage is to eliminate the human factor (mistake) at the time of the transaction and to be able to act by 24 hours daily. Algorithmic trade takes place on the basis of an adopted strategy. Algo trading institutions also employ professional staff in mathematics and engineering.

The aim of such teams is to develop the most effective strategy. We should also mention the role of so-called self-learning algorithms (machine learning/machine learning) used for algo trading.

Wikipedia is characterized by machine learning as an area of artificial intelligence devoted to algorithms that improve automatically through experience, or exposure to data.

Algorithmic trading using machine learning collects data on transactions concluded, then analyses them on an ongoing basis by drawing conclusions and thus makes the most favourable transaction decision.

3. High-Frequency Trading (HFT)

According to the preamble to the MiFID Directive, a specific sub-group of algorithmic commerce is a high-frequency algorithmic trade, whereby the transaction system immediately analyses data or signals from the market and then sends or updates a large number of orders in a very short time in response to the results of that analysis. In particular, high-frequency algorithmic trading may include elements such as triggering, generating, redirecting and execution, which are determined by a system without human participation for each individual transaction or order, having a short term to determine and liquidate position, a high daily portfolio turnover, a high ratio of orders to transactions, intraday and end-of-day, equal to or close to a flat position[12].

By formulating this more simply, algorithmic trading with high frequency allows, using a computer program, to include a large number of transactions in microseconds[13]. This gives an advantage over other market participants, which, under favourable conditions, manages to conclude several to several transactions within a minute.

Most of the profit from one HFT transaction is a fraction of a penny, but taking into account the large number of transactions made over the period one seconds on a scale one hours may already be significant amounts[14].

one from examples of the use of HFT is the application of the so-called arbitration strategy. It involves using the price difference of the financial instrument in question as two or more markets.

Thanks to this strategy, the algorithm captures the difference in price, which usually takes up to a few seconds and then makes a favorable transaction. The speed with which the algorithm works allows it to use a specific "weather window" which is generally not available to traditional traders.

The strength of HFT algorithms is their speed, far beyond human capabilities.

An interesting form of HFT strategy is a strategy called News read algorithms . These algorithms use statistical methods and text mining techniques 15 estimate the impact on the market of upcoming news.

In this strategy, the algorithm uses available publications on macroeconomic data or information coming from social networks (Twitter, LinkedIn, Facebook).

Such an algorithm, after receiving a signal, uses its natural advantage, which is the speed of operation, which allows it to enter into within seconds of publication of the information the most affordable transaction.

The remaining market participants, not using HFT algorithms, usually include these late transactions (in relation to HFT), even when the price of the instrument has changed.

Technical progress has enabled high-frequency transactions and the evolution of business models. The execution of high-frequency transactions is conducive to the location of market participants' seats within a short distance of the trading venue's execution systems. Algorithmic transactions or high frequency algorithmic trading techniques may, like any other form of transaction, be susceptible to certain forms of behaviour prohibited under the MAR Regulation[16].

A high-frequency trading strategy, which may constitute financial market manipulation, should include:

• 1) quote stuffing consists of the multiple rapid introduction and withdrawal of a large number of orders to flood the market. In this way, the price of the instrument in question, at short intervals, alternately increases and decreases slightly.
• The purpose of such a strategy is to delay the decision by other algorithm users. In this way, HFT, using the delay no, can introduce a larger and more significant strategy 17 ;

1. layingering and spoofing – are algorithms using which the market is “dipped” by buying and selling orders, which shifts orders from other market participants. This algorithm contains transactions in the opposite direction of the original order, allowing it to get a better price. If this strategy is used, there may be a rapid price collapse in the markets called Flash Crash.[18].

Both strategies, although fairly similar, are aimed at other market participants. The strategy of quote stuffing is aimed solely at algorithm users, while layingering and spoofing are aimed at non algorithm users.[19].

4. Summary

The popularity of HFT trade is growing in Poland, but we are far from yet to the United States or the countries of the European Union. Technological progress, the speed of Internet connections, as well as digitisation and automation promote the development of algorithmic commerce.

New ways of manipulating the capital market using new technologies will also emerge with technological progress. The European Union is trying to tackle the negative consequences of such practices, but it seems that EU legislative activity is only a late response.

_______________________________________