Cryptocurrencies as a subject of financial fraud

PURPOSE: The main purpose of this paper was to identify the current scope of research on cryptocurrencies as a subject of fraud. Detailed research questions related to the determination of contemporary trends of the conducted research and the definition of potential opportunities for further investigation of this topic. One of the questions also concerned identifying the most common crimes committed using cryptocurrencies. METHODOLOGY: The study is based on a systematic literature review (SLR) of 57 publications available on the Scopus database. A bibliometric and descriptive analysis of selected literature items was carried out. Then, vital thematic clusters were separated, and an in-depth content analysis was performed. FINDINGS: The detailed bibliometric and descriptive analysis showed that cryptocurrencies as a subject of financial fraud are generally a new area of scientific research, although it is developing quite intensively. The relatively small number of publications, compared to other similar areas, also indicates that this topic has not yet been explored widely by scientists, and many different research trends can be created in it. Ultimately, the following key research areas were identified: types of cryptocurrency fraud, crime detection methods, risks related to blockchain technology, money laundering, and legal regulations regarding cryptocurrencies. It was also possible to identify that money laundering is currently the most common fraud. However, it has been pointed out that the second most frequent fraud is financial pyramids based on the Ponzi scheme. IMPLICATIONS: The paper clearly presents the main research trends on using cryptocurrencies in criminal activities. At the same time, it was emphasized that, compared to other research areas, this topic is relatively new. Therefore, there is a wide possibility of exploring not only existing but also undiscovered research trends. In addition, key types of fraud in economic practice have been identified, which is particularly important for financial market participants. It was clearly indicated which transactions bear the highest risk. It is also worth paying


INTRODUCTION
Bitcoin was the first cryptocurrency to be created by Satoshi Nakamoto in 2009. Its introduction to the economic market completely revolutionized many existing mechanisms related to the financial market. Some changes related to entire foundations, i.e., defining new economic concepts or changing the current perception of selected macro-and microeconomic processes.
The key terms in this context are "cryptocurrency" and "blockchain" related to new technologies. In short, it can be concluded that the first is a generic term for a virtual or digital currency that takes the form of coins or tokens. Cryptocurrencies use blockchain technology (Al-Saqaf & Seidler, 2017). In turn, blockchain is defined as a chained data structure that combines blocks of data and information in chronological order and records the blocks in encrypted form as a distributed ledger that cannot be tampered with or forged. It uses timestamps to identify and record each transaction, so the data are traceable, thereby preventing irreversible modifications to data or information (Lu, 2019). So cryptocurrencies do not require a central authority to validate and settle transactions. Instead, they use only cryptography (and an internal incentive system) to control transactions and manage the supply. Payments are validated by a decentralized network (Gandal, Hamrick, Moore, & Oberman, 2018).
From the very beginning, the essence of blockchain and related cryptocurrencies has been the subject of research by scientists. It is possible to identify many research trends by analyzing the potential application areas of the new technology. Blockchain can be used to decentralize the financial system (Chen & Bellavitis, 2020;Patel, Migliavacca & Oriani, 2022;Sánchez, 2022), to create new forms of nonfungible token (NFT) investments (Regner, Schweizer, & Urbach, 2019), and finally to implement smart contracts (Cong & He, 2019;Hughes, Park, Kietzmann, & Archer-Brown, 2019;Rozario & Vasarhelyi, 2018), which are systems that automatically control digital assets according to arbitrary prespecified rules. There is also an increasing trend of its mass application in accounting and financial reporting of enterprises (Schmitz & Leoni, 2019;Pimentel & Boulianne, 2020;Kokina, Mancha, & Pachamanova, 2017). Therefore, further intensive changes in economic systems caused by this invention should be expected.
Thousands of new cryptocurrencies have been introduced in recent years. It is estimated that over 5,100 such assets are currently in operation (Goforth, 2021). The scale and variety of abuses related to it are also growing. The US Federal Trade Commission published a report that presented the latest data on the scale of fraud in the cryptocurrency market (FTC, 2021). It shows that from October 2020 to the end of March 2021, almost 7,000 people fell victim to virtual currency fraud, which resulted in a total loss of more than $ 80 million. For comparison, this sum is nearly 1000% higher than the amount recorded in the corresponding period a year earlier. The report presented by FTC also shows that the median value of the losses suffered was $ 1,900. Global data on the topic can be found in Chainanalysis reports. According to the latest information, the value of frauds and scams in the crypto-assets market in 2021 amounted to $ 14 billion, i.e., almost twice as high ($ 7.8 billion) as the year before (Chainanalysis, 2022). It is also worth paying attention to the scale of money laundering by use of this market. Cybercriminals laundered $ 8.6 billion worth of cryptocurrency in 2021. That represents a 30% increase in money laundering activity over 2020. Cybercriminals have laundered over $ 33 billion worth of cryptocurrency since 2017.
As presented above, the problem is beginning to grow and it significantly affects the security of economic transactions on a global scale. This has been pointed out by both ordinary market participants as well as the governments of individual countries. We are seeing a dynamic increase in global financial flows that are not under any effective control. In connection with the above, it is highly desirable to identify the degree of development and the scope of the current research on the cryptocurrency market in the context of their use for criminal activities. A preliminary analysis of the literature on the subject indicates an existing research gap. There are some studies that present an analysis of the literature on the subject to date (Trozze et al., 2022). The authors have made a scoping review of academic research and grey literature on cryptocurrency fraud. When selecting scientific positions, the Google Scholar search engine was used, and for the remaining publications -the Google search engine. The main purpose of the study was to identify the types of crimes committed with the use of cryptocurrencies. Hence, only those items that contained a description of at least one example of fraud (as of November 2020) were analyzed. As a result of this work, 29 different types of cryptocurrency fraud included in scientific publications and 32 types discussed in the grey literature were distinguished. However, it should be mentioned that only the identification of the types of fraud using cryptocurrencies is insufficient. It is worth making a broader analysis of the literature on the subject in order to define also other areas of research in this field. This article takes that perspective.
The main purpose of the study is, therefore, to identify the current scope of research on cryptocurrencies as a subject of fraud. It will also allow the definition of potential opportunities for further investigation of this topic. To the best of the author's knowledge, it is one of the first studies showing the results of a systematic literature review (SLR) combined with a bibliographic analysis and an in-depth analysis of the content of publications in this field. Detailed research questions are presented later in the study. This article is structured as follows. The next part deals with the general theoretical background of the subject, with particular emphasis on research conducted in similar areas. Then the research methodology is presented. In this section, research questions are posed. The process of selecting publications for their systematic review is explained in detail, as well as the approach to bibliometric and descriptive analysis. The basic parameters used to distinguish thematic clusters are also presented in this part of the study. The following section shows the results of a systematic review of the literature, including the findings of an in-depth analysis of the content of individual publications. They provide the basis for a discussion on the context of the specific research questions. The last part contains conclusions and presents potential directions for further research in this field.

METHODOLOGY
This study adopted a systematic approach to conducting a literature review to minimize bias and lend scientific value to its results. Systematic literature review (SLR) is a widely recognized scientific method used in social sciences, including management, economics, and finance (Hiebl, 2021;Simsek, Fox, & Heavey, 2021;Sharma & Bansal, 2020). According to the guidelines included in the literature on SLR, the study was divided into the following stages (Jesson, Matheson & Lacey, 2011;Booth, Sutton, & Papaioannou, 2016): • defining research questions; • searching for the literature; • selection of publications using exclusion and inclusion criteria; • preparation of the final database; • bibliometric analysis; • content analysis; • discussing the results.
At the beginning, three main research questions were defined that set the direction and scope of the systematic literature review, especially in content analysis. The following questions were asked: RQ1) What are the current state and the primary considerations of research relating to cryptocurrencies as a subject of fraud? RQ2) What are the most common crimes committed with the use of cryptocurrencies? RQ3) What could be the future research trends related to cryptocurrencies and financial fraud?
A systematic literature review is crucial in responding to RQ1 and RQ3. In turn, RQ2 is also related to the practical implications of the subject of the study. From the methodological point of view, one of the essential elements of a systematic literature review is an appropriately conducted process of selecting a research sample. The individual steps of eliminating and including in the final set of publications should be based on clear criteria and performed in the correct order. Selected literature items indicate different sampling activities (Sharma & Bansal, 2020;Denyer & Tranfield, 2009;Gaur & Kumar, 2018). However, as a rule, three standard main stages can be distinguished in them: • identification -it consists in determining a potential group of publications relating to a predefined research problem (Vassar et al., 2017;Booth et al., 2016); • screening -application of various criteria for inclusion and exclusion of selected items to the final research sample, relating mainly to the substantive content, including also the qualitative assessment of the publication based on content analysis (Booth et al., 2016;Pussegoda et al., 2017;Briner & Denyer, 2012); • final review sample -determining the definitive set of literature items on the subject being the basis for a detailed analysis from the point of view of the research questions posed. In this respect, there are several guidelines for the minimum dataset size. In the context of the analyzed issue, the minimum sample size should be 50 items (Short, Sharma, Lumpkin & Pearson, 2016;Hiebl, 2021).
A diagram of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was used to present the different stages of determining the final set of scientific publications covered by the analysis. This model is one of the most frequently used tools that regulates the sequence of actions performed during the sampling process (Pussegoda et al., 2017;Page et al., 2021;Liberati et al., 2009). Scopus was selected as the key database for the systematic literature review. Before deciding on the choice of the final database, trial tests were also carried out for other databases of scientific publications, especially the Web of Science and ProQuest. However, preliminary results of searching these databases using comparable criteria indicated a smaller number of publications and they included many duplicates. Therefore, it was decided to use Scopus, where the scope of the publication was the largest. The time range of the published scientific items was not limited due to the relatively new subject of scientific research, i.e., cryptocurrencies. An interesting aspect was also the identification of the oldest publications in this field.
The first stage of searching the database and selecting items was determining the keywords appropriate for the research subject. This collection includes cryptocurrencies and crypto assets (crypto *), bitcoin, Ethereum, fraud, crime, scam, and abuse. The first four keywords generally refer to cryptocurrencies and their two most popular and longest-functioning types in the market. The following four keywords are a combination of the most common terms related to financial fraud in the literature. The "Article title, abstract, keywords" area was selected as the reference database for the search. As a result of the database search, 841 publications meeting these criteria were identified.
Then, the subject area had to be narrowed down due to its substantive nature. In this regard, two sites were selected: "business, management, and accounting" and "economics, econometrics, and finance." The scope of the publication was 106 items. Another criterion was to narrow the area of analysis to four types of documents: "article," "conference paper," "book chapter," and "book." The database identified 102 publications. Of these, all articles still in print were discarded, and the focus was solely on the completed items. As a result of the database search, 95 scientific publications were finally included in the collection, and the full search criteria were as follows: The next stage was verifying the titles and abstracts of all 95 bibliographic items to determine which of them relate to the research questions posed. The mainstream research was supposed to concern cryptocurrencies in the context of fraud committed. At this stage, a complete double analysis of titles and abstracts was performed to eliminate the risk of confusion. Thirtyeight publications were rejected. For research purposes, the final collection was 57 literature items. The eliminated publications mainly concerned the possible innovative applications of blockchain technology related to the cryptocurrency market in other areas of life (medicine, education).
The summary of the entire process of selecting the research sample is the following PRISMA diagram presenting the various stages of the elimination of bibliographic items (Figure 1).

IDENTIFICATION -records identified from Scopus n = 841
SCREENING -records excluded by general subject area (only "business, management and accounting" and "economics, econometrics and finance") n = 735 -records excluded by type of document (only "article", "conference paper", "book chapter" and "book" -final version) n = 11 -records excluded by double content analysis of titles and abstracts n = 38 FINAL REVIEW SAMPLE -number of records for in-depth analysis n = 57 In the next stage, bibliometric and descriptive analyses were made to present the primary data on the literature on this subject. The key verification criteria concerned publication trends in particular years, types of these publications, the most popular journals, authors, their affiliation, and countries of origin. The publications were also analyzed in terms of their levels of citation. Using the tools built into the Scopus database, a ranking of the most frequently cited literature items from the collection included in the final analysis was prepared. In this way, it was possible to identify critical publications that were most often included by other authors dealing with similar issues.
The last element was the in-depth content analysis. The starting point for organizing the substantive criteria of this analysis was the verification of potential clusters. Therefore, a network analysis was performed using the VOSviewer 1.6.18 Software. The data was extracted directly from Scopus, including all necessary information (author, title, abstract, keywords, publication year, affiliation, etc.), and then imported to VOSviewer to create the co-occurrence network to identify the main aspects of the discussion. The most crucial element in this respect was the analysis of the co-occurrence network of keywords to distinguish clusters. The key parameters used to define the network of connections are presented in Table 1. Subsequently, each literature item within the individual clusters was read, and an in-depth content analysis was performed. A narrative approach was used during this verification, and significant substantive findings relating to the research questions were presented.

RESULTS
The first element is a detailed bibliometric and descriptive analysis. The subject of cryptocurrencies in the context of crime is relatively new compared to other research areas related to management, economics, and finance. Figure 2 presents the evolution of the number of publications in this field. As mentioned above, no filters related to time constraints were assumed when searching the Scopus database. The results, therefore, present the full range of literature on the subject.
One of the analysis's most interesting elements was identifying the oldest publications on cryptocurrencies used as a potential fraud tool. The above chart shows that in 2014 one such item was published. After that, only individual publications were identified over the next several years. It was only in 2018 that higher growth dynamics can be observed -eleven such items were published then. A similar trend continued in the following years (12, 15, and 12 publications in 2021). The chart does not present data from 2022, as the information does not include the whole year and it would distort the conclusions from the comparative analysis. This implies that after the first cryptocurrency was introduced to the market, at least a few years had to pass before it became the object of interest to scientists. It is also clear that in the early years, cryptocurrencies were not identified as potential financial crime tools.

Figure 2. Number of publications in respective years
Another aspect of bibliometric verification is the publication type and the most famous journals. The data show that among the entire group of 57 literature items, the most significant number is of scientific articles (36) and publications in conference materials (14). Trace amounts refer to chapters in books or complete monographs. A summary of these data is presented in Figure 3.

Figure 3. Types of publications
The publications appeared in a total of 42 different journals and conference materials. The analysis showed that articles in this field are most often published in the Journal of Money Laundering Control, which immediately suggests the essence of the research issue and the most common type of fraud in the context of cryptocurrency trading. A total of 10 publications were identified in this journal. The following items are publications in conference materials relating to the IEEE International Conference on Blockchain and Cryptocurrency, ICBC 2020, and four more journals. Two elements of the literature on the subject were identified. The rest are single publications in 36 different journals. It is immediately noticeable that there is one top place dealing with the issue of cryptocurrencies in the context of economic fraud. Table 2 presents the primary sources of publications in this field. Considering the number of publications issued by individual authors, it is difficult to identify a leading scientist specializing in this subject. Only five people with two publications on using cryptocurrencies in financial frauds can be distinguished -Esoimeme, Falker, Moore, Teichmann, and Wronka. It is worth pointing out that all of their scientific articles are about money laundering with crypto-assets and have been published in the Journal of Money Laundering Control. Most researchers dealing with this subject come from the United States, Great Britain, and China. Due to the large dispersion of publications issued by individual authors, it is also impossible to indicate specific research centers related to the affiliation that would play a leading role in scientific research on this subject. Table 3 presents the ranking of publications from their citation point of view. It includes all items for which more than 10 citations were identified. In total, for the entire set of 57 items, 921 citations were established in other scientific publications, of which 11 articles did not receive any citations in the analyzed period of 2018-2022.  When analyzing the above results, three leading items of the most significant substantive importance with more than 100 citations should be identified. They mainly relate to manipulating the bitcoin exchange rate or its use to build a financial pyramid (Ponzi scheme).
The next stage of the systematic literature review is the in-depth content analysis of the collection of publications from the point of view of achieving the main goal and the research questions posed. The starting point for this analysis was the identification of potential substantive clusters. As mentioned earlier, VOSviewer Software was used to create the co-occurrence network using all keywords to identify the main aspects of the publications.
As part of the selection process presented in the methodology section, a total of 322 keywords were identified, of which only 10 met the assumed criteria. The number of occurrences and total link strength for the most important keywords are presented in Table 4. Generally, three clusters focused on the following keywords were identified: • bitcoin -electronic money, cryptocurrencies; • blockchain -crime, chromium compounds, Ethereum, block -chain; • cryptocurrency -money laundering.
The results are presented in Figure 4, which shows network visualization. In addition, the density visualization in Figure 5 was also included to provide a complete presentation of the selected clusters.

Figure 5. Clusters of the authors' keywords
Source: own study using VOSviewer 1.6.18.

Cluster "Bitcoin"
Bitcoin as the oldest cryptocurrency and various aspects related to its functioning, is by far the most frequently discussed topic in the literature on the subject. Many authors emphasize that the perception of bitcoin has significantly evolved and is now a normally functioning asset in financial markets. Therefore, Teo and Low (2018) pointed out that it is necessary to redefine the concept of "money" as an asset and define its protection principles. They showed various legal aspects related to defining this concept and the resulting risks in investment practice. It was especially emphasized that the main threat in this regard is hacking. A large part of the publication shows examples of possible illegal use of bitcoin for various crimes, including money laundering (Esoimeme, 2018;Bartoletti et al., 2018;Barth et al., 2020;Broadhead, 2018, van Wegberg et al., 2018. Esoimeme (2018) indicates that bitcoin creates hitherto unknown opportunities for marketing funds from illegal sources, much more significant than traditional money transfers. For example, the Mavrodi Mondial Movement (MMM) pyramid scheme operating in recent years in Nigeria and the risks associated with new payment methods are given.
Ponzi scheme-based financial pyramids are one of the most commonly used frauds in the context of bitcoin, with the longest tradition in the market. It is noted by Bartoletti, Pes, and Serusi (2018), Zhang, Kang, Dai, Chen, and Zhu (2021), Wang, Cheng, Zheng, Yang, and Zhu (2021). They build a network of investors, where the profits paid to the first participants of the system come from payments made by subsequent investors and not from the funds generated by the system. The authors emphasize that immediately after the introduction of bitcoin in 2009, there were signs of building financial pyramids with its use. At the same time, they proposed various techniques for detecting bitcoin addresses directly related to Ponzi schemes, allowing early identification of this type of fraud. The method involves experimenting with different machine learning algorithms and evaluating their effectiveness using standard validation protocols and performance metrics. In turn, Wang, Cheng, Zheng, Yang, and Zhu (2021) proposed a method for detecting pyramid schemes based on oversampling Long Short-Term Memory. Account features and code features are extracted from contract call information and contract codes, and the two components are combined to detect Ponzi scheme smart contracts.
Analyzing bitcoin addresses and their associated transaction types is also of interest to Lin, Wu, Hsu, Tu, and Liao (2019). They point out that the ability to identify addresses associated with criminal activities is becoming the most critical issue in the cryptocurrency network. They experimented Marta Gancarczyk, Małgorzata Kutera, Óscar Rodil-Marzábal (Eds.)

/ Cryptocurrencies as a subject of financial fraud
with building a classification model for detecting abnormality of bitcoin network addresses. These features include various high orders of moments of transaction time, which summarizes the transaction history in an efficient way. This allows the addresses associated with the scams to be identified. A modern tool for detecting suspicious bitcoin accounts was also proposed by Sun, Xiong, Yiu, and Lam (2019), who developed the BitVis system. With it, cryptocurrency investors can easily filter transactions on demand, interact with trading networks to find helpful information, and analyze the behavior of bitcoin accounts. The mechanism may also be successfully used by authorities regulating financial markets.
A tool popularly known as honeypot (Torres, Baden & State, 2020) can also play a similar role. This particular trap is aimed at detecting attempts at the unauthorized use of the system or obtaining data. Most often, it consists of a computer, data, and a separate area of the local network, which pretends to be a real network but are isolated from it and adequately secured. From the outside, this construction looks like it contains information or a resource that could be a potential target of a cybercriminal. Another publication that presents the potential possibilities of preventing fraud with bitcoin is the study on the innovative solution proposed by Toyoda, Ohtsuki, and Mathiopoulos (2018). Scientists implemented a multi-faceted scheme for identifying services based on bitcoin addresses by analyzing the history of transactions. It allows distinguishing seven significant services: regular exchange, faucet, gambling, investment scam, marketplace, mining pool, and mixer. The model provides 72% accuracy and it has been tested on over 26,000 bitcoin addresses. In turn, Lorenz, Silva, Aparício, Ascensão, and Bizarro (2020) conducted experiments to detect illegal activity in a set of bitcoin transactions. They studied the detection ability of the machine learning model and proved that unsupervised anomaly detection methods have poor results.
Interesting research in the context of bitcoin has also been presented by Barth, Herath, and Xu (2020). These scientists were looking for answers as to whether, and to what extent, ethical aspects affect the valuation of cryptocurrencies. To this end, they measured the intensity of the use of ethical and unethical words in the discussion of bitcoin on Twitter and its valuation. They discovered that the frequency of an unethical discussion about bitcoin is negatively associated with its price. In contrast, the frequency of an ethical debate is positively associated with its price.

Cluster "Blockchain"
The publications belonging to this cluster mainly concern blockchain as a new technology covering, among other things, the cryptocurrency market and they draw attention to various associated risks. It is assumed that this solution completely revolutionized the existing digital world and brought an entirely new perspective on its security, flexibility, and efficiency (Srivasthav, Maddali, & Vigneswaran, 2021). On the one hand, it is emphasized that blockchain allows for a completely different dimension of transactions or exchange of goods and services. However, its further development depends to a large extent on regulatory changes protecting against cybercrimes and financial frauds (Ahram, Sargolzaei, Daniels & Amaba, 2017).
The WannaCry ransomware attack that took place in May 2017 was given as an example of a new type of crime involving blockchain-based cryptocurrency payment transactions (Turner, McCombie, & Uhlmann, 2019). It was a global hacking attack that involved computers running the Microsoft Windows operating system by encrypting data and demanding ransom payments in the bitcoin cryptocurrency. At the same time, the authors developed a model for collecting and analyzing data related to inflows and outflows of bitcoin-related ransomware transactions. Bitcoin transactions form graph networks and enable the construction of a target network model for collecting, analyzing, and sharing intelligence with multiple stakeholders. It would therefore be possible to counter such attacks more quickly and effectively in the future. Karapapas, Pittaras, Fotiou, and Polyzos (2020) draw attention to the increased risk of hacker attacks using blockchain technology. The authors clearly showed how technology could be used to launch ransomware campaigns as a service. They proved that criminals could transact with related parties and victims without revealing their identity and with multiple privacy guarantees. The scale of cyber-attacks in cryptocurrency trading and the use of technology was also the subject of research by Caporale, Kang, and Spagnolo (2020). They thoroughly analyzed hacking attacks on the four most popular cryptocurrencies. They confirmed their significant negative financial consequences and, at the same time, pointed to the need to increase research in this field. They considered the precise understanding of the mechanisms of cyber-attacks to be crucial in the fight against this phenomenon.
With the development of blockchain and the cryptocurrency market, the scale of abuse related to the simple theft of these assets has also increased. Only the tools used by criminals have changed. These scams operate on visually similar but seemingly unrelated websites advertised by malicious social media accounts. With the help of such websites and social media accounts, they Marta Gancarczyk, Małgorzata Kutera, Óscar Rodil-Marzábal (Eds.) / Cryptocurrencies as a subject of financial fraud often perpetrate fraud or act as phishing sites. For example, Phillips and Wilder (2020) analyzed selected data online and based on blockchain technology. Using the clustering technique, they developed a typology of prepayment and phishing scams. It turned out that the same entities carried out very similar scams in their online activities and using blockchain.

Cluster "Cryptocurrency"
The vast majority of publications in this field see the problem of using cryptocurrencies for various crimes, including primarily money laundering. Levin, O'Brien, and Zuberi (2015) explicitly point out that until recently, the bitcoin market was considered a "virtual Wild West for drug dealers and other criminals." At the same time, they pointed out that the support for this currency is constantly growing, and it has become a global virtual asset. The regulations governing this market do not keep up with the practice and seem unclear. The authors cite examples of American administrative proceedings against operators of platforms on which cryptocurrency trading is carried out and analyze the current state of legal regulations in this field in the USA.
The issue of appropriate regulation was also raised by Irwin and Dawson (2019), who specifically dealt with the law of payment methods. The authors identified the current legal status in Australia, Europe, and America and, at the same time, indicated potential limitations in their application on a global scale. In addition, they highlighted the ineffectiveness of the implemented solutions, which also have a negative impact on the possibility of prosecuting criminals. One of the reasons they mentioned is the lack of a legal, universally binding definition of bitcoin.
It is also emphasized that countries that give up cash transactions entirely are not much less vulnerable to money laundering crimes (Rivera, 2019;Hendrickson & Luther, 2022). In this case, virtual transactions, including those related to cryptocurrencies, are used on a larger scale. After all, popular cryptocurrencies like bitcoin are close substitutes for cash. In addition, they offer a higher level of financial anonymity and thus allow transactions with a lower risk of detection than traditional digital payments. Consequently, all efforts to eliminate cash from circulation strongly drive criminals towards cryptocurrencies.
Experts indicate that a substantial restriction of trading in cryptocurrencies is not the solution for the future either, because they appeared as a natural consequence of the intensive development of technology. However, it is essential to introduce global legal regulations limiting their criminal use (Al-Tawil & Younies, 2020). Liechtenstein is quite an active country in this context (Teichmann & Falker, 2020;. Particular guidelines have recently been introduced regulating this market ("The Liechtenstein Blockchain Act"), preventing money laundering above all. New regulations were also introduced relatively quickly in Malta (Buttigieg & Sapiano, 2020). Teichmann and Falker (2020) also presented specific methods used by people involved in money laundering using crypto assets. The qualitative research included 10 presumed money-laundering people and 18 anti-money-laundering experts.
Quite an exciting publication was prepared by Dupuis and Gleason (2021). The authors presented the possibilities and limitations of the cryptocurrency market as a place for money laundering. They performed an in-depth analysis of the currently available exchange mechanisms of these assets in light of the existing legal regulations. The illegal use of cryptocurrencies was investigated through Kane's regulatory dialectical paradigm and it eventually identified six potential tools used by criminals.
A similar topic was taken up by Lui and Ryder (2021). They classified the mechanisms of using cryptocurrencies in financial crimes and analyzed the relevant legal provisions in Great Britain. There has also been an attempt to identify the current loopholes in the regulatory systems that are most often exploited by fraudsters. The authors emphasized that, despite the efforts of the Financial Action Taskforce, the legal system does not keep up with the development of technology, and harmonized global actions are needed in this regard.
Potential money laundering techniques using cryptocurrencies were also presented by . He classified the most common fraud mechanisms and patterns and highlighted the changing cryptocurrency market that brings new opportunities for fraud. The author also analyzed the extent to which EU and national regulations can counteract this phenomenon, bearing in mind the security of the financial market. In verifying domestic law, he dealt mainly with the legal provisions in force in Germany, Great Britain, and Switzerland. Findings suggested that relatively lenient laws exist in Switzerland and Germany, while Great Britain has the most stringent regulations.
The directions of changes in the law in the context of the security of cryptocurrency trading were also presented by Fletcher, Larkin, and Corbet (2021). They performed an in-depth analysis of the regulations in the American market. The authors distinctly indicated that bitcoin and other crypto-assets should be classified as a technology with financial components and regulated as a part of the growing FinTech industry. In turn, Riley (2021) reviewed current Chinese law, with particular emphasis on the new Chinese Cryptography Law. Marta Gancarczyk, Małgorzata Kutera, Óscar Rodil-Marzábal (Eds.) / Cryptocurrencies as a subject of financial fraud

DISCUSSION
The bibliometric and descriptive analysis summary presented above made it possible to partially answer the first research question regarding the state of scientific publications on cryptocurrencies as a subject of fraud. There has been an increased interest in this topic for several years.
The detailed content analysis of the literature items made it possible to indicate the main research trends, identify the most frequently committed frauds with the use of cryptocurrencies and define future research directions, which are closely related to the research questions posed. The starting point was the identification of three keyword-based clusters. The verification of the publications included in individual groups indicated some thematic specialization.
Within the bitcoin cluster, the most significant number of studies presenting various types of crimes committed with its use and possible techniques for detecting these abuses were identified. Some publications also referred to the need to redefine critical concepts related to cash turnover or ethics. Cluster "blockchain" clearly focuses on new technology and the resulting risks. On the other hand, the group of studies in the field of cryptocurrencies mainly refers to issues related to money laundering and changes in international and national legal regulations regarding cryptocurrency trading. Figure 6 summarizes the identified vital research trends.

Types of cryptocurrency fraud
Crime detection methods

Money laundering
Legal regulations regarding cryptocurrencies The two main types of economic crime related to the cryptocurrency market are money laundering and financial pyramids based on the Ponzi scheme. All authors point out that this market offers new and unprecedented possibilities for transferring funds from illegal sources. This is facilitated by certain anonymization of transactions and the lack of clear legal regulations. As soon as the formalized framework for the organized cryptocurrency trading market, including mainly bitcoin at first, emerged, criminals found new opportunities for money laundering. This type of abuse is by far the most common in the cryptocurrency context (Esoimeme, 2018;Levin et al., 2015;Rivera, 2019;Hendrickson & Luther, 2022;Bartoletti et al., 2018;Barth et al., 2020;Broadhead, 2018;van Wegberg et al., 2018;Dupuis & Gleason, 2021;.
However, attention should be paid to the second, quite a strong trend of publications on the use of the cryptocurrency market to build Ponzi schemes (Esoimeme, 2018;Bartoletti et al., 2018;Zhang et al., 2021;Wang et al., 2021). The crimes of the financial pyramid have been known in the market for many years and they have always aroused a lot of emotions, mainly due to the scale of actions of selected fraudsters and the wide range of victims. The very name of the type of fraud comes from Charles Ponzi, an Italian immigrant living in the United States. In 1920, he built the first financial pyramid (based on the international reply coupons IRC). Since then, this type of fraud has systematically appeared in the market. One of the largest frauds of this type in the economic history of the world is the financial pyramid of Bernard L. Madoff, a well-known American stock exchange player. The number of victims exceeded fourteen thousand people and the losses were estimated at tens of billions of dollars (Kutera, 2016). The cryptocurrency market offers new opportunities in this regard, although the essence of the crime has remained unchanged. It was presented in detail by Wang, Cheng, Zheng, Yang, and Zhu (2021), who described the fraud mechanism using the example of PlusToken. However, there are more examples: the OneCoin-based pyramid operating in or BitConnect (2016. Another area of research is the methods of detecting cryptocurrency scams. Most researchers here focus on various ways of verifying bitcoin addresses and identifying those that bear the hallmarks of criminal activity (Bartoletti et al., 2018;Lin et al., 2019;Toyoda et al., 2018). Other proposals relate to machine learning models (Wang et al., 2021;Lorenz et al., 2020) or completely original solutions (Sun et al., 2019;Torres et al., 2020). In this context, everyone emphasizes that the capabilities of blockchain technology can also contribute to a more effective fight against economic crime related to cryptocurrencies. This trend of research also applies to the IT sector, where you can see a growing number of publications describing the use of so-called smart contracts. Therefore, it seems that the subject of creating various application tools supporting fraud detection in the blockchain environment will be a separate and stringent research stream.
The analysis of the content of publications regarding various risks arising from the use of blockchain technology and their potential impact on the cryptocurrency market showed that the main problem is hacker attacks. Selected studies presented examples of such situations and identified the scale of financial losses (Turner et al., 2019;Broadhead, 2018). The most significant illegal acquisitions of cryptocurrencies as a result of imperfect information systems took place, for example, on Mt. Gox, where the size of the financial damage was estimated at $ 473 million. Other examples include the hacking attacks on the Bitfinex exchange in August 2016 (total losses amounted to $ 72 million), PolyNetwork in August 2021 ($ 600 million), and Zaif in September 2018 ($ 62 million). Some publications in this area also presented more technical aspects related to the actual carrying out of attacks and analyzed the main IT tools used in the crime (Karapapas et al., 2020;Caporale et al., 2020;Phillips & Wilder, 2020).
The last highlighted research area is that of legal regulations regarding cryptocurrencies and their impact on the security level of this market. The analysis covers both global and national levels. The authors identified the current state of the law and the desired directions of its changes. The regulations applied in the United States (Levin et al., 2015;Fletcher et al., 2021), Great Britain (Lui & Ryder, 2021;, China (Riley, 2021), Australia (Irwin & Dawson, 2019), Germany and Switzerland  and in smaller countries such as Liechtenstein (Teichmann & Falker, 2020; and Malta (Buttigieg & Sapiano, 2020). In some cases, broader international comparative analyzes were carried out, which allowed for more profound conclusions. In general, attention was drawn to the urgent need to develop and implement some global standards regulating the cryptocurrency market. The rules applied at the national level are insufficient to protect investors fully. Crypto-asset transactions, by their nature, involve transnational cash flows.

CONCLUSION
The main purpose of this paper was to identify the current scope of research on cryptocurrencies as a subject of fraud. Ultimately, 57 publications were selected for the systematic review of the literature. The detailed bibliometric and descriptive analysis showed that it is generally a new area of scientific research, although it is developing quite intensely. The relatively small number of publications compared to other similar areas also indicates that this topic is not yet explored widely by scientists, and many different research trends can be created within it.
In turn, an in-depth analysis of the content made it possible to find answers to the specific research questions. They mainly referred to identifying the most critical trends in the current research on cryptocurrencies in the context of financial fraud and the definition of potential opportunities for further investigation of this topic. The starting point in this part of the study was the identification of three thematic clusters and more detailed areas of analysis within them. Ultimately, the following key research trends were identified: types of cryptocurrency fraud, crime detection methods, risks related to blockchain technology, money laundering, and legal regulations related to cryptocurrencies. One of the questions also concerned the practical implications of the research area, namely identifying the most common crimes committed with the use of cryptocurrencies. These include money laundering and financial pyramids based on the Ponzi scheme.
The contribution of this study is threefold. It is one of the first research papers showing the results of a systematic literature review (SLR) combined with a bibliographic and in-depth analysis of the content of publications in this field. This is all the more important as the scale of crimes involving cryptocurrencies is growing yearly, which is also mentioned in this study. Secondly, the key types of fraud have been identified that, at the same time, cause the most significant financial loss. This allowed for the establishing of directions for further research, which have profound practical implications for market participants. The most important issues that should be included are: • desired changes in the field of international and national legal regulations regarding cryptocurrency trading, which on the one hand, would increase the security of investors, but at the same time, would not inhibit the natural development of new solutions emerging along with the dynamic technological development; • in-depth research on identifying possible types of fraud committed using cryptocurrencies, mainly to build effective mechanisms to combat these phenomena. In this respect, cooperation of specialists in various fields, for example, financiers and IT specialists, would be desirable; • creating and analyzing various application tools supporting fraud detection in the blockchain environment.
However, there are also limitations to this study. Regarding the research methodology, the Scopus database does not allow the analysis of all available publications related to the topic (including studies only in paper form). Moreover, only items published in English were taken into account during the selection of articles. Various reports prepared by organizations dealing with the analysis of the cryptocurrency market or institutions responsible for Marta Gancarczyk, Małgorzata Kutera, Óscar Rodil-Marzábal (Eds.) / Cryptocurrencies as a subject of financial fraud shaping legal regulations in this field were also not taken into account. The main goal of the article was closely related only to scientific publications. It is also worth mentioning the time limit. The selection of the items in the literature on the subject was made as of July 2022. Therefore, the analysis did not cover the latest publications, which may be important in the context of the dynamic development and changes that have taken place on the cryptocurrency market in the recent period of time. Despite this, the author believes the study will be a helpful resource for current and future scholars interested in addressing the most critical connections between cryptocurrencies and financial crimes.