Journal of Information Technology (2017)

ª 2017 Association for Information Technology Trust All rights reserved 0268-3962/17
www.palgrave.com/journals

Research Article

A taxonomy of financial market

manipulations: establishing trust
and market integrity in the financialized
economy through automated fraud
detection
Michael Siering1, Benjamin Clapham1, Oliver Engel1, Peter Gomber1
1
Goethe University Frankfurt, Theodor-W.-Adorno-Platz 4, 60323 Frankfurt, Germany

Correspondence:
M Siering, Goethe University Frankfurt, Theodor-W.-Adorno-Platz 4, 60323 Frankfurt, Germany.
Tel: +49 (0)69 798 34683;
Fax: +49 (0)69 798 35007;
E-mail: siering@wiwi.uni-frankfurt.de

Abstract
Financial market manipulations represent a major threat to trust and market integrity in
capital markets. Manipulations contribute to mispricing, market imperfections and an
increase in transaction costs for market participants and in costs of capital for issuers.
Manipulations are facilitated by increased transaction velocity, speculative trading and
abusive usage of new trading technologies, i.e., they are directly linked to financial sector
changes that drive financialization. Research at the intersection of financialization and IS
might support regulatory authorities and market operators in improving market surveillance
and helping to detect fraudulent activities. However, confusing terminology is prevalent on
financial markets with respect to different manipulation techniques and their
characteristics, which hampers efficient fraud detection. Furthermore, recognizing
manipulations is challenging given the large number of information sources and the vast
number of trades occurring not least because of high-frequency traders. Therefore,
automated market surveillance tools require a comprehensive taxonomy of financial
market manipulations as a basis for appropriate configuration. Based on a cluster analysis
of SEC litigation releases, a review of the latest market abuse regulation and academic
studies, we develop a taxonomy of manipulations that structures and details existing
manipulation techniques and reveals how these techniques differ along several
dimensions. In a case study, we show how the taxonomy can be utilized to guide the
development of appropriate decision support systems for fraud detection.
Journal of Information Technology (2017). doi:10.1057/s41265-016-0029-z

Keywords: financial market manipulation; market surveillance; taxonomy; fraud detection;

decision support

Introduction
he financial crisis triggered an intensive debate about financial activity does not fulfill its main purpose, i.e., to

T the role of the financial sector in national and

international economies. There is widespread agree-
ment that weaknesses in financial regulation and supervision
contributed to the crisis and that the strongly increased
serve the needs of the real economy by providing efficient
financial products and services. Financial activities are viewed
as an end in itself to enable financial players to generate
excessive profits based on highly complex financial products
 A taxonomy of financial market manipulations
(e.g., over-the-counter (OTC) derivatives) and a tremendous
increase in turnover on financial markets. These trends are
mirrored in the term financialization, which is defined as ‘‘the
increasing dominance of financial actors, markets, practices,
measurements and narratives, at various scales, resulting in a
structural transformation of economies, firms (including
financial institutions), states and households’’ (Aalbers, 2016:
3). Key drivers of financialization are deregulation, the focus
on the shareholder value concept and, last but not least, the
proliferation of information technology (IT) (Lagoarde-
Segot, 2016). New trading technologies such as algorithmic
trading and high-frequency trading (HFT) lead to massive
increases in turnover velocity on financial markets because
they enable sophisticated market participants to generate
significant total profits based on small profits per trade and
to instantaneously react to profitable situations in a highly
complex and fragmented trading environment (Gomber
et al., 2016).
Furthermore, since financial markets have existed, market
participants try to generate profits from deceiving others
(Leinweber and Madhavan, 2001). For instance, stock prices
are manipulated to pretend a trend (Comerton-Forde and
Putnins, 2011), or false and misleading information is
disseminated to cause market reactions and to profit from
other market participants’ actions. However, in recent years,
the magnitude of financial market manipulations and their
effect has increased continuously. Technology-savvy market
participants, such as HFTs, are perceived to exploit their
technological advantage to perform deceptive tactics (Biais
and Woolley, 2011; Cumming et al., 2012) at the expense of
other market participants or to contribute to extreme and
unexpected market price fluctuations as witnessed in the May
6, 2010, Flash Crash (Easley et al., 2011). This event brought
these new trading technologies to the forefront of the
financialization debate and triggered far-reaching regulatory
initiatives both in the USA and in Europe to monitor or even
curb this form of speculative trading, which is accused of
focusing solely on market microstructure noise rather than
contributing to market efficiency by incorporating relevant
information into asset prices.
In addition to this increased risk of speculative mispricing
and potential market manipulation due to new technologies,
numerous concrete scandals and accusations in the context of
market abuse and manipulation were recently uncovered.
The manipulation of the fixings in the foreign exchange
markets over several years (Financial Conduct Authority
(FCA), 2014) and the Libor rate manipulation scandal
(Financial Services Authority (FSA), 2012) revealed massive
rigging activities in the area of financial benchmarks that led
to fines of more than $10 billion until 2016 (Vaughan, 2016).
Politicians, regulators and the public thus push for increased
transparency and massive efforts in the area of surveillance to
curb market abuse and manipulation and to re-establish trust
in the financial sector.
Due to the massive increase in transactions triggered by
speculative trading, an enormous amount of data must be
analyzed to generate alerts based on overnight-batch pro-
cesses that can be investigated in detail to identify potential
and actual market manipulations. Because order and trading
data can hardly be analyzed manually, automated fraud
detection systems are required that provide alerts of potential
M. Siering et al
market manipulations. Previous research has already shown
that appropriate information systems (IS) can help to
identify fraudulent content (Abbasi et al., 2010; Zahedi
et al., 2015) and communication behavior (Zhou et al.,
2004). Nevertheless, although the necessity of decision
support systems (DSSs) that support market surveillance
authorities in the field of financial fraud detection appears
obvious, few academic studies exist in this particular context.
One essential prerequisite for building DSSs that are able to
assist with the identification of potentially fraudulent behav-
ior is an in-depth domain and data understanding (Fayyad
et al., 1996) that prevents improper system configurations and
might lead to the discovery of wrong and invalid patterns.
Although a clear distinction between different manipula-
tion techniques is a prerequisite for efficient market surveil-
lance, confusing terminology is prevalent with respect to
manipulation techniques and their main characteristics;
existing classifications (e.g., Allen and Gale, 1992) neither
provide a comprehensive view of differences and similarities
nor cover the newly emerging forms of manipulation in
conjunction with HFT. Furthermore, they do not link to the
configuration of fraud detection systems.
To close this research gap and to follow the call of Ngai
et al. (2011) concerning the automated detection of fraud in
the field of financial markets, we develop a novel and
comprehensive taxonomy of market manipulations by (1)
analyzing current cases of prosecuted market manipulations
based on a cluster analysis of litigation releases published by
the US Securities and Exchange Commission (SEC), (2)
performing a structured literature review to include all
manipulation techniques studied by academics and (3)
completing the taxonomy based on an analysis of current
financial market regulation. That is, we include the European
Market Abuse Regulation (MAR) for multi-country coverage
of market manipulations. Thus, we follow the methodology
proposed by Nickerson et al. (2013) and build a taxonomy of
market manipulations, considering both the differences and
similarities of various market manipulation techniques and
new types of manipulations related to HFT strategies in
particular.
We contribute to research at the interface of financializa-
tion and IS by (1) systematically documenting and organizing
new and established forms of market manipulation that are
based on traditional and new (potentially speculative) trading
technologies, such as HFT, in a market manipulation
taxonomy and (2) evaluating this taxonomy for the auto-
mated detection of market manipulations based on sophis-
ticated information systems. Furthermore, we provide specific
guidelines on how to derive a DSS configuration to detect
fraudulent market behavior based on a systematically derived
taxonomy, thereby contributing to research in IS. The results
are highly relevant for regulators and market surveillance
authorities because the systematic and automated detection of
market manipulation prevents abusive behavior and con-
tributes to re-establishing trust in financial markets.
This paper is structured as follows. Section ‘‘Research
context: financial market manipulations and their detection’’
presents related work on financial market manipulations and
fraud detection. Section ‘‘General research approach’’ briefly
explains the general research methodology including the
taxonomy approach. Section ‘‘Analysis of financial market
 A taxonomy of financial market manipulations
manipulations based on SEC litigation releases’’ provides first
insights into market manipulations based on a cluster analysis
of litigation releases. Insights into our structured literature and
regulatory review are provided in section ‘‘European market
abuse regulation and literature review’’. Section ‘‘Developing a
taxonomy of financial market manipulations’’ presents the
taxonomy development. Section ‘‘Evaluation of the taxonomy’’
evaluates our taxonomy and shows in a case study how the
taxonomy can be applied in the field of fraud detection.
Finally, section ‘‘Conclusion’’ summarizes our findings.
Research context: financial market manipulations and their
detection
Manipulation of financial markets is as old as the markets
themselves and has been studied extensively in the academic
literature both from an economic and from a legal perspec-
tive. To avoid manipulative actions and to ensure capital
markets’ integrity, exchanges and regulatory authorities
engage in market surveillance to detect manipulative behav-
ior (Leinweber and Madhavan, 2001).
Because manipulations can occur in different forms and
categories, there is no generally accepted definition of
financial market manipulation or financial fraud (we do
not distinguish between market fraud and manipulation, and
both terms are used interchangeably in the following).
Therefore, we follow a broad definition of financial market
manipulations; first, market manipulations encompass the
manipulation of financial disclosures accompanied by large
accounting scandals such as Enron and Worldcom, because
such scandals hamper the efficient functioning of markets
(Akhigbe et al., 2005; Jones, 2011). Second, investor confi-
dence is impeded by investment manipulations such as the
Ponzi scheme set up by Bernard Madoff (Pozza et al., 2009;
Rapoport, 2012). Third, insider trading is another form of
market manipulation, drawing on information asymmetries
(Arshadi, 1998; Allen and Ramanan, 1995). Fourth, there is a
variety of different fraudulent trading activities aiming at
financial market manipulation that we also consolidate under
this umbrella term. According to Cumming and Johan (2008:
456), such ‘‘trading practices … distort prices and enable
market manipulators to profit at the expense of other
participants, creating information asymmetries’’. Forms of
such market manipulations are, for example, ‘‘Pump and
Dump’’ techniques in which the manipulator releases false-
positive information concerning a company to profit from
the subsequent price increase (Sabherwal et al., 2011) or
‘‘Marking the Close’’ in which the fraudster tries to manip-
ulate the closing price of a stock (Hillion and Suominen,
2004). These manipulations have a severe effect on trust in
financial markets and on investors, who can lose substantial
parts of their investment, e.g., due to ‘‘Pump and Dump’’
fraud (Hanke and Hauser, 2008). A detailed overview of
market manipulations addressed in academic studies and of
international market abuse regulation is provided in our
literature and regulatory review (‘‘European market abuse
regulation and literature review’’ section).
Due to the increasing utilization of IT in particular, which
is also one main driver of financialization (Lagoarde-Segot,
2016), the risk of market manipulations increases due to
novel forms of manipulation arising from technologies such
as HFT. Furthermore, the chance of manually detecting
M. Siering et al
market manipulations is lowered due to increasing amounts
of data to be analyzed. To increase trust, the prevention of
market manipulations is of high importance, which requires
appropriate IS for fraud detection. Research on IS indicates
that automated detection systems are able to find patterns in
datasets to identify fraudulent Web content (Abbasi et al.,
2010; Zahedi et al., 2015). Additionally, related classification
methods might be used to detect deceptive content in
computer-mediated communication (Zhou et al., 2004).
Nevertheless, previous research has also shown that an in-
depth domain understanding is essential for finding an
appropriate system configuration (Fayyad et al., 1996).
In the field of financial market manipulations, Allen and
Gale (1992) are among the first who provide a structured
overview and categorization. They differentiate between
actions that manipulate the actual or perceived value of an
asset (action-based manipulation), the release of false or
misleading information to move stock prices in the desired
direction (information-based manipulation) and attempts to
manipulate by buying and selling a stock in a specific manner
(trade-based manipulation). However, the taxonomy devel-
oped by Allen and Gale (1992) is not sufficiently detailed for
use as an input for an automated market surveillance system
because it does not address a potential system configuration.
Moreover, due to technological enhancements, new forms of
manipulation have evolved that are not covered by this
taxonomy. For example, algorithmic traders and in particular
the subset of HFTs might try to manipulate stock prices
based on order submissions and their subsequent deletions to
profit from their low-latency access to markets and market
data.
Manipulation is a serious threat to the efficiency and
liquidity of trading venues and undermines the trust of
market participants (Kyle and Viswanathan, 2008). There-
fore, detection of market manipulation/financial fraud is
essential to ensure the orderly function of markets and to
provide a level playing field for all participants. Distinguish-
ing fraudulent financial data from authentic data is one of the
most difficult tasks regulators face (Ngai et al., 2011). In the
USA, the SEC is responsible for market manipulation
detection and prosecution. Self-regulatory (non-governmen-
tal) institutions such as the Financial Industry Regulatory
Authority (Davis, 2007) are also active in this field. In the
European Union (EU), the MAR, which has been in force
since 2014 and has been applied since July 2016, lists
activities that are defined as financial market manipulation.
There are a variety of tools that can be used by regulators
to detect market manipulation (Wheeler and Aitken, 2000).
Of special interest for this study is the use of intelligent
systems for decision support to identify suspicious behavior
(Baker, 2005; Humpherys et al., 2011). Although there is a
large research stream about the empirical evidence of
manipulations and their effects on financial markets, little
research addresses the detection of financial market fraud by
using an automated DSS (Comerton-Forde and Putnins,
2011).
One approach to detecting manipulation is proposed by
Aggarwal and Wu (2006), who analyze 142 cases of stock
market manipulation. They find that potentially informed
parties and market makers are likely to be manipulators and
show that prices of manipulated stocks rise through the
manipulation period and fall thereafter. Comerton-Forde
 A taxonomy of financial market manipulations
and Putnins (2011) developed a measure of the probability of
closing price manipulation in a given market for a stock.
Furthermore, Ledgerwood and Carpenter (2012) derived a
theoretical framework for classification and detection of
market manipulation. Other work focuses on peer group
analyses using unsupervised learning to detect suspicious
behavior (Kim and Sohn, 2012) or data mining techniques to
detect fraudulent financial statements (Kirkos et al., 2007).
However, no generally accepted basis for the development of
an automated financial fraud detection system exists.
Furthermore, none of the mentioned approaches contains a
means to detect market manipulation conducted by HFTs or
other market participants using a low-latency infrastructure.
With this study, we therefore respond to the call for research
by Ngai et al. (2011) and contribute to the scarce literature
on financial fraud detection by proposing a taxonomy of
market manipulations that is of fundamental importance to
configure automated fraud detection systems.
General research approach
The primary goal of this study is to develop a taxonomy of
financial market manipulations. Taxonomy development has
been studied intensively in the social sciences. A detailed
overview is provided by Bailey (1984). We apply the
methodology developed by Nickerson et al. (2013), which
is accepted in the field of IS, and adapt it by including
different sources for input, i.e., currently existing market
manipulations, as shown in Figure 1.
We identify market manipulation techniques based on
three sources from academia and international regulation:
SEC litigation releases to determine fraudulent financial
market activities that are prosecuted in the USA, manipu-
lation techniques that are described in the MAR and a
structured literature review on financial market manipula-
tions. Additionally, we rely on the European Benchmark
Regulation (European Parliament and Council, 2016) as a
source for fraudulent actions aimed at the manipulation of
market benchmarks. As a result, we determine the different
forms of market manipulation which are then used together
with a first taxonomy developed by Allen and Gale (1992) as
a starting point to derive our taxonomy, which shows
mutually exclusive and collectively exhaustive characteristics.
The already existing dimensions included in the taxonomy
are determined based upon established category definitions
in the academic literature. New dimensions are built drawing
on insights from the structured literature review and
international market abuse regulation. An evaluation and
discussion of the proposed taxonomy follows.
Figure 1 General research approach for taxonomy development.
M. Siering et al
Analysis of financial market manipulations based on SEC
litigation releases
SEC litigation releases and dataset acquisition
Litigation releases (LRs) report the enforcement actions of
the SEC. Among other information, a release can include
arraignments, a subpoena (e.g., defendants in a civil court
explain themselves), an announcement of investigation, a
final judgment (e.g., penalty or disgorgement payments are
announced), a trading halt (e.g., to protect the unsuspecting
public from manipulation) or public service announcements.
LRs represent a prominent source to analyze market manip-
ulations that are prosecuted by the SEC.
We downloaded all LRs from the SEC homepage from
December 29, 1995, to July 22, 2013, resulting in 8093
releases. To avoid data inconsistency and noise, we cleaned
the obtained dataset by excluding years that were only partly
covered due to the introduction of the database in 1995/1996
(and the incomplete year 2013), LRs that did not refer to
Rule 10b-5 of the SEC, which covers the ‘‘employment of
manipulative and deceptive devices,’’ and large scandals and
Ponzi schemes because they tend to involve large numbers of
defendants, claimants, and different courts, which leads to
redundancy in the dataset and might influence the clustering
process (Coffee, 2005). Nevertheless, Ponzi schemes are later
included in the taxonomy because they represent a form of
financial market manipulation. Table 1 depicts the data
processing in detail.
Our final dataset covers 4340 releases. Table 2 provides
summary statistics for the documents in the total raw dataset
and for the documents that are used for the analysis.
Clustering approach
To derive insights into which general categories of manip-
ulative behavior are prosecuted by the SEC, the LRs are
clustered. Therefore, the raw data must be transformed in a
form that is readable and processable by an unsupervised
clustering algorithm (Rokach and Maimon, 2005). The
objective behind the unsupervised clustering approach is to
reveal novel patterns and to explore an unknown database
without the requirement of specific knowledge about the
particular structure of the LRs.
The documents are preprocessed before the clustering
algorithm is applied. Therefore, they are reduced to their
fundamental parts, and a term frequency measure is applied
to create a word vector of the documents. First, the
documents are tokenized by applying a split after every
non-letter character (e.g., commas, semicolons and spaces)
(Rehman et al., 2013). After that, uppercase letters are
 A taxonomy of financial market manipulations
transformed into lowercase letters to homogenize the words.
Next, one-letter tokens and stop-words are excluded from the
list of tokens. These exclusions are non-informative tokens
such as articles, pronouns and adverbs (Fatudimu et al.,
2008). To enhance the meaningfulness of the individual
tokens, a Porter-Stemmer is used (Porter, 2006). By reducing
tokens to their common stem, the total number of different
tokens is reduced, which also reduces computational com-
plexity. The advantage of this algorithm is context awareness
and simplicity of the suffix removal rules (Willett, 2006). In
the last step, we calculate TF-IDF scores for every token
(Jones, 1972), and every token is assigned a number that
determines its weight in the document (term frequency) and
in the overall corpus of documents (inverse document
frequency).
Following the pre-processing, we apply a clustering
algorithm. Clustering is the assignment of objects to groups
with high intra-similarity and low inter-similarity; many
algorithms exist to solve this optimization problem, each
having its own specific field of application (Tabrizi et al.,
2013). The LRs are expressed in natural unstructured
language, which has important consequences for usable
algorithms and measures because LRs have different lengths
and content structures (Losiewicz et al., 2000). Therefore,
algorithms must be used that can process high-dimensional
datasets and normalize the documents to avoid ambiguous
results (Ye, 2011). For the task of finding patterns in high-
dimensional datasets, the k-means algorithm combined with
a suitable similarity measure is the algorithm of choice
(Willett, 1988). The k-means algorithm divides a population
of N-Dimensional Objects (here LRs) into k sets by
optimizing the within-class variance and calculating the
means of the attributes (centroids) within the cluster
(MacQueen, 1967). Practically, k random vectors are created,
and the similarity to the objects is measured, with the most
similar objects being grouped together. This process is
repeated until the within-class variance is optimized or the
predefined maximum number of iterations is reached. The
Table 1 Excluded documents from the raw litigation release database
Cleansing measure Number of excluded
documents
Filter 1995,1996 and 2013 722
No 10b-5 reference 1845
Filter Ponzi schemes 632
Filter scandals (e.g., Enron, 554
Worldcom)
Final database 4340
Table 2 Summary statistics for the analyzed litigation releases
Measure Documents per year
Raw
Average 425
Median 438
Minimum 122
Maximum 619
SD 120
M. Siering et al
assignment of a mean vector (centroid) to every cluster
makes this algorithm especially useful for document cluster-
ing because the mean vector can easily be interpreted and
applied by human reasoning (MacQueen, 1967).
Clustering process and evaluation of the different cluster models
We calculate 15 different models with numbers of clusters
from 2 to 16. There is no commonly agreed evaluation
technique or measure for determining the most suitable num-
ber of clusters and clustering model (Aliguliyev, 2009).
Clustering always requires both quantitative (e.g., perfor-
mance measures) and qualitative (e.g., human reasoning or
expert knowledge) criteria to determine the true number of
clusters and to evaluate the found patterns (Frank, 1994).
Therefore, we apply both qualitative criteria and the X-means
algorithm to find the optimal number of clusters (Bouwman,
1983; Pelleg and Moore, 2000).
Table 3 shows an excerpt of the results of the clustering
algorithm for the optimal cluster number of four. The
table presents the twelve top scoring tokens for each cluster
together with the corresponding cosine similarity (TF-IDF
value). In comparison with the other models developed, the
reported cluster model with four clusters appears to be the
most suitable from a qualitative point of view. Different from
the models with fewer clusters, there is an information gain
by adding up to four clusters to describe the dataset.
Furthermore, four categories of market manipulation are
revealed: financial instrument manipulation, accounting fraud,
investment fraud and insider trading. Each fraudulent activity
is characterized by the respective relevant keywords that
appear most often within a cluster as shown in Table 3.
Because the tokens ‘‘stock’’ and ‘‘manipul’’ are of high
importance for the first cluster, it is labeled the cluster of
financial instrument manipulation (Allen and Gale, 1992).
The other tokens in this cluster fit to financial instrument
manipulation, e.g., penny stocks are common targets of
manipulative actions (see token ‘‘penni’’) (Bartels, 2000), and
several manipulation techniques such as ‘‘Pump and Dump’’
depend upon the Internet (see token ‘‘internet’’) to dissem-
inate false messages (Sabherwal et al., 2011). The second
cluster is determined to represent accounting fraud because it
clearly includes LRs that cover the misstatement or over-
statement of accounting figures such as revenue (see token
‘‘revenu’’) or financial figures (see token ‘‘financi’’) (Gerety
and Lehn, 1997). This statement also holds true for the other
tokens in this cluster such as ‘‘audit,’’ ‘‘fiscal’’ and ‘‘report.’’
Cluster 3 covers manipulation with respect to funds of
investors (see token ‘‘fund’’ and ‘‘investor’’) and is named
investment fraud cluster (Carroll, 2006). Investment fraud is
regularly conducted by asset managers or financial advisors
Word count
Cleaned Raw Cleaned
271 3694 1678
274 3363 1537
208 819 406
371 48306 7767
41 1721 700
 A taxonomy of financial market manipulations M. Siering et al

Table 3 Result of the clustering process with four centroid vectors (k = 4) showing tokens, corresponding ranks, TF-IDF scores and a representative litigation release
for each cluster

Rank Cluster 1 Cluster 2 Cluster 3 Cluster 4

Financial instrument manipulation Accounting fraud Investment fraud Insider trading
(1283 LRs) (840 LRs) (1351 LRs) (866 LRs)
Token TF-IDF Token TF-IDF Token TF-IDF Token TF-IDF
1 Stock 0.039 Revenu 0.050 Fund 0.064 Insid 0.065
2 Judgment 0.030 Report 0.045 Invest 0.059 Trade 0.056
3 Internet 0.027 Record 0.041 Investor 0.053 Tip 0.043
4 Manipul 0.027 Audit 0.040 Advis 0.040 Option 0.040
5 Florida 0.024 Quarter 0.039 Llc 0.036 Share 0.040
6 Penni 0.024 Fiscal 0.037 Client 0.035 Profit 0.037
7 Offer 0.023 Financi 0.034 Capit 0.032 Acquisit 0.036
8 Defend 0.023 Account 0.034 Asset 0.031 Purchas 0.034
9 Share 0.022 Auditor 0.030 Manag 0.030 Stock 0.033
10 Final 0.022 Aid 0.030 Defend 0.025 Merger 0.033
11 Investor 0.020 Year 0.029 Return 0.024 Nonpubl 0.032
12 Enter 0.020 Offic 0.029 Bank 0.023 Illeg 0.030
LR LR 21423 (2010) LR 17346 (2002) LR 22545 (2012) LR 17645 (2002)

(see tokens ‘‘asset,’’ ‘‘manag’’ and ‘‘advis’’) using the legal form
of a limited liability company (see token ‘‘llc’’) and deceiving
clients by disbursing false returns (see tokens ‘‘client’’ and
‘‘return’’) (Ionescu, 2010). The last cluster indicates that the
most relevant technique for the cluster is insider trading (see
tokens ‘‘insid’’ and ‘‘trade’’) (Roddenberry and Bacon, 2011).
Insider trading often occurs around mergers and acquisitions
(see tokens ‘‘merger’’ and ‘‘acquisit’’); the manipulator tries to
profit from non-public information (see tokens ‘‘profit’’ and
‘‘nonpubl’’) (Allen and Ramanan, 1995).
Three tokens, i.e., ‘‘investor,’’ ‘‘stock’’ and ‘‘share,’’ appear
in two of the four clusters. However, this appearance does
not harm the validity of the clustering process because both
financial instrument manipulation and investment fraud are
conducted either by investors or to deceive investors.
Likewise, financial instrument manipulation aims at the
manipulation of stock, respectively, share prices while market
participants conducting insider trading try to benefit from
changes in stock (share) prices based on non-public
information.
Because these four clusters are meaningfully distinguish-
able, a minimum number of four clusters appears appropri-
ate. By adding one more cluster, the investment fraud cluster
is split. This split, however, does not help to describe the
dataset better but instead creates fuzzy assignments because
five of the twelve tokens assigned to cluster three (see
Table 3) are also relevant for the new cluster of the model,
making it difficult to find a clear and distinguishable label for
both clusters. This finding and the analysis of the rest of the
centroid vectors (six to fifteen) leads to the conclusion that
k = 4 is the optimal number of clusters from a qualitative
information gain standpoint (Bouwman, 1983).
After having analyzed the optimal cluster number from a
qualitative standpoint, a k-means algorithm is applied as a
quantitative measure. Specifically, we apply the X-means
algorithm that computes all cluster models (similar to the k-
means approach) from a predefined upper to lower boundary
and assigns scores based on the Bayesian Information
Criterion to every model (Pelleg and Moore, 2000). When
a score is computed for every model in the predefined range,
the model with the best score determines the optimal number
of clusters to partition the dataset from an information
criterion point of view. The quantitative approach supports
the chosen model of four clusters.
In addition to the four clusters of financial market
manipulations that we identified based on SEC LRs, a deeper
analysis of the centroid vector for the cluster of financial
instrument manipulations reveals more fine-grained manip-
ulative actions such as ‘‘Pump and Dump’’ (indicated by the
token ‘‘pump’’ on rank 47 of the centroid vector). A more
detailed analysis of different manipulation techniques is
provided in the ‘‘European market abuse regulation and
literature review’’ section to enhance and complete the
taxonomy as described in the ‘‘Developing a taxonomy of
financial market manipulations’’ section.
European market abuse regulation and literature review
Analysis of European regulation
With our taxonomy of financial market manipulations, we
want to provide a comprehensive, multi-country overview
on fraudulent behavior in financial markets. Therefore, we
rely on not only US legislation but also inclusion of
manipulation techniques that are considered in the MAR.
By examining both US and European jurisdictions, we have
considered all manipulations in the two largest financial
markets. Different from the LRs, no in-depth textual
analysis is necessary for the MAR because manipulation
techniques can be extracted from the articles defined in the
regulation.
The MAR is intended to ensure the integrity and smooth
functioning of European financial markets with the goal of
increasing investor confidence (European Parliament and
 A taxonomy of financial market manipulations M. Siering et al

Council, 2014). The regulation precisely describes different manipulation techniques mentioned in at least two sources,
activities that are considered manipulative behavior in and we categorized the different fraudulent activities
Articles 8 and 12. Furthermore, we determined abusive according to clusters based on the analysis of SEC LRs.
activities related to market benchmarks from the European
Benchmark Regulation, effective as of 2016 (European
Parliament and Council, 2016). The list of market manip- Developing a taxonomy of financial market manipulations
ulations extracted from the MAR and the European Bench-
mark Regulation is presented together with the additional Taxonomy development methodology
manipulation techniques found in academic studies in Table 5 Developing a suitable taxonomy is important for almost
(see next section). every discipline to understand the underlying mechanics and
to break down complex topics into smaller parts (Miller and
Literature review of financial market manipulations Roth, 1994). The emphasis when developing a taxonomy of
In addition to incorporating regulation and prosecution of manipulation techniques should be on awareness of the
market manipulations, we conduct a systematic literature effects on the market and the identification of touch points
review to identify manipulations that might not yet be with the market environment (Kumar and Langberg, 2009).
considered in regulatory documents. This emphasis leads to the development of effective coun-
A systematic literature review shall provide a compre- termeasures or detection techniques that help to make
hensive view on the topic in question (Denney and financial markets more efficient and less likely to be
Tewksbury, 2013) and exhaustively include all related manipulated (Cumming and Johan, 2008). To achieve this
studies to reveal the current state of knowledge in the goal, comprehensibility and extendibility are highly impor-
research field under investigation (Rowley and Slack, 2004). tant so that the taxonomy can be adapted to changing
In this context, an effective literature search relies on market conditions and new manipulation techniques (Ngai
meaningful keywords and backward and forward reference et al., 2011).
searches (Levy and Ellis, 2006) that are especially fruitful To the knowledge of the authors, no concise, robust,
because the keyword search is rather unlikely to fully yield comprehensive, extendible and explanatory taxonomy of
all relevant literature concerning the topic (Webster and manipulation techniques can be found in the financial
Watson, 2002). market manipulation literature (Nickerson et al., 2013; Kyle
We applied both techniques to find all manipulation and Viswanathan, 2008). There only exist simple taxonomies
techniques that are addressed in academic research. We for market manipulations. For instance, Allen and Gale
searched in peer-reviewed journals in reputable databases for (1992) distinguished action-, information- and trade-based
scientific publications from 1990 onwards, as shown in manipulation.
Table 4. Keywords in our search were ‘‘stock market Our taxonomy development is based on the method
manipulation’’, ‘‘stock market fraud’’, ‘‘financial market proposed by Nickerson et al. (2013). Nickerson et al. (2013:
manipulation’’ and ‘‘financial market fraud’’. Applying back- 338) define taxonomies as ‘‘[…] systems for grouping objects
ward and forward reference searches, we manually added of interest in a domain based on common characteristics.’’
additional relevant publications not found in the keywords The algorithm follows an iterative approach. Objects are
search. discovered and grouped according to defined characteristics
Table 5 presents all of the financial market manipulations and dimensions.
that we identified based on SEC LRs, the European MAR For the taxonomy, the objects are the manipulation
and the systematic literature review. We only included techniques, and the characteristics should be related to our
meta-characteristic, the market environment, also encom-
passing the interaction of the manipulation with the market
Table 4 Results of the systematic literature review environment. This interaction involves both market partic-
ipants (issuers, intermediaries and investors) and market
Database Number of Publications explicitly information such as prices and trading volumes (Ledgerwood
hits in the covering financial and Carpenter, 2012; Hellwig, 1980).
database market manipulations The algorithm starts with the most comprehensive char-
acteristic, and each (sub) characteristic should be a logical
ABI/INFORM database 125 3
consequence (Nickerson et al., 2013). After each iteration, a
ProQuest
taxonomy table (see Table 6) and a taxonomy formula [see
ACM 35 3
Eq. (1)] are derived until the ending conditions are met.
AIS library 14 0   
Business Source Premier 252 19 Tm ¼ Di ; i ¼ 1; . . .; njDi ¼ Cij ; j ¼ 1; . . .; ki ; ki  2
Emerald Fulltext 18 6
IEEE Xplore Digital 3 2 ð1Þ
Library Tm is the taxonomy after iteration m with a set of n
Science Direct 43 14 dimensions Di, each consisting of ki mutually exclusive and
Springer-Link journals 53 4 collectively exhaustive characteristics Cij. The markers (‘‘X’’)
JSTOR 65 6 in the table indicate the classification of the objects. The
Manually added 17 17 exemplary taxonomy shown in Table 6 is described by the
Total 625 74 following Eq. (2):
 A taxonomy of financial market manipulations M. Siering et al

Table 5 Identified market manipulation techniques

Cluster (SEC Manipulation Description References

LRs) technique
Accounting Fraudulent Accounting fraud means the willful Gerety and Lehn (1997), Bonner et al.
fraud financial misrepresentation of the financial health of a (1998), Pagano and Immordino
statements firm by disclosure violations and improper (2012)
accounting. There exist different types of
accounting fraud (e.g., earnings inflation,
fictitious transactions and accounts payable
fraud).
Investment Benchmark Collusion of banks or other intermediaries to Abrantes-Metz et al. (2012), Fouquau
fraud manipulation influence and manipulate fundamental and Spieser (2015), MAR Art. 12
(with official reference prices or interest rates that are (1)(d), European Parliament and
fixing) officially fixed at a certain point in time (e.g., Council (2016)
LIBOR manipulation).
Benchmark Collusion of banks or other intermediaries MAR Art. 12 (1)(d), European
manipulation aimed at the manipulation of fundamental Parliament and Council (2016)
(no official reference prices or interest rates (e.g.,
fixing) commodity benchmarks such as oil or gold
prices).
Ponzi scheme A Ponzi scheme is a fraudulent investment Shleifer and Vishny (1997), Ionescu
operation in which the operator, an (2010), Rapoport (2012)
individual or organization, consecutively
raises external funds and pays returns to
investors from new capital raised from later
investors, thereby creating an illusion of high
returns.
Insider trading Insider trading Insider trading means that investment Roddenberry and Bacon (2011),
decisions are based on relevant non-public Arshadi (1998), Allen and Ramanan
information (e.g., executives trading before (1995), MAR Art. 8 (1)
dividend announcements).
Financial Pump and dump The fraudster buys relatively unknown or low- Sabherwal et al. (2011), Bartels (2000),
instrument valued stocks, disseminates false-positive MAR Art. 12 (1)(c), MAR Art. 12
manipulation information to lure other investors into (2)(d)
buying the stock, which leads to increasing
stock prices (the ‘‘pump’’). In the next step,
the fraudster sells the shares to realize
substantial profits before the stock price
reverts to its normal low level (the ‘‘dump’’).
Short and distort Short and distort describes the manipulation Leinweber and Madhavan (2001), Vila
technique of an investor entering a short (1989), MAR Art. 12 (1)(c), MAR
position in a specific company’s stock, Art. 12 (2)(d)
disseminating wrong negative information
about the company and buying back the
stock after its decline, thereby making
substantial profits.
Cornering The manipulator corners the market of a Møllgaard (1997), Golmohammadi
(squeezing) security by obtaining large quantities of the et al. (2014), Lomnicka (2001), Allen
security, thereby gaining a price-controlling et al. (2006), Cooper and Donaldson
market position due to the shortage of (1998), Jarrow (1992), Putnins
supply. Third parties that, for example, must (2012), MAR Art. 12 (2)(a)
fulfill contracts based on short positions are
then forced to buy the security at inflated
prices (squeezing).
Advancing the Increasing the bid for a security to artificially Cumming and Johan (2008), Klein
bid increase its price. et al. (2012), MAR Art. 12 (1)(a)(ii)
Reducing the ask Reducing the ask for a security to artificially Cumming and Johan (2008), Klein
decrease its price. et al. (2012), MAR Art. 12 (1)(a)(ii)
 A taxonomy of financial market manipulations M. Siering et al

Table 5 Continued

Cluster Manipulation Description References

(SEC technique
LRs)
Matched orders Transactions in which colluding traders enter Fischel and Ross (1991), Lomnicka (2001),
matching buy and sell orders simultaneously Cumming and Johan (2008)
to feign an active market to lure other
investors into buying the security, thereby
causing a price increase.
Painting the tape Painting the tape refers to a trader engaging in Cumming and Johan (2008), Carhart et al.
a series of publicly reported transactions to (2002)
give the impression of trading activity and
price movements. As for matched orders, the
objective is to attract other investors buying
the security, leading to higher stock prices.
However, collusion of several traders is not a
necessary characteristic of painting the tape.
Wash sales Wash sales describe a manipulation technique Thel (1993), Lomnicka (2001), MAR Annex 1
of fictitious transactions performed by one (A) (c)
trader without change of actual ownership
because the same party is buyer and seller to
create a record of rising prices and the
illusion of an active market.
Capping A practice in which the price of a security Cumming and Johan (2008), Putnins (2012)
(pegging) underlying an option is manipulated shortly
before the option’s expiration date to prevent
a rise/decline in price of the security so that
the previously written call/put option will
expire worthless, thereby protecting the
option premium initially received.
Marking the Placement of buy orders with high limits/sell Comerton-Forde and Putnins (2011),
close/open (also orders with low limits to boost/suppress the Cumming and Johan (2008), Hillion and
banging the closing (opening) price of a security. This Suominen (2004), MAR Art. 12 (2)(b),
close) manipulation technique is often used to CFTC (2016), Karz and Wagner (2006)
influence performance measures that are
based on securities’ closing prices (less
frequent: opening prices). The technique is
also called Banging the Close when a trader
transacts a large amount of an asset during
the closing period to benefit from another,
even larger position, e.g., in an option or
other derivative.
Front running Front running refers to brokers or market Cataldo and Killough (2003), Chaturvedula
makers making use of their private et al. (2015)
information about incoming order flow by
buying or selling a security in advance of other
parties’ large trades, thereby profiting from the
price movement that follows the large trades.
Churning Churning describes the activity of a broker Shapiro et al. (2012), Cumming and Johan
engaging in excessive buying and selling of (2008)
securities in a client’s account to create
higher commissions, disregarding the client’s
interests.
Scalping Scalping refers to manipulative action of Hazen (2010), Tripp (1963), MAR annex 1
investment advisors purchasing a security (B) (b)
shortly before recommending that security to
third parties without disclosing their
position. The fraudster profits from the rise
in the price following the recommendation
and immediately sells off his shares.
 A taxonomy of financial market manipulations M. Siering et al

Table 5 Continued

Cluster Manipulation Description References

(SEC technique
LRs)
Spoofing Spoofing refers to the placement of limit orders Lee et al. (2013), Biais and Woolley (2011),
that are not intended to be executed but to MAR Art. 12 (1)(a)(i)
mislead other investors with respect to the
demand or supply of a security. The
manipulator later submits his real order,
thereby profiting from the price change
resulting from his spoofing activity.
Pinging Pinging describes the submission of small IIROC (2012), Scopino (2015)
marketable orders without an intention to
trade to detect large hidden orders (abusive
liquidity detection) with the intention of
benefiting from that information.
Quote stuffing Quote stuffing is a technique employed by IIROC (2012), Biais and Woolley (2011),
HFTs that involves the placement and Easley et al. (2011), MAR Art. 12 (2)(c)(i)
immediate cancelation of a large number of
orders in an attempt to flood the trading
system with excessive messages. Quote
stuffing can create information arbitrage
opportunities for HFTs due to increased data
latencies for other market participants.
Ramping HFTs enter buy orders at successively Cumming and Johan (2008), MAR Art. 12
increasing prices to lead other investors to (2)(c)(iii)
perceive an active interest in a security.
Layering (order Layering describes a strategy in HFT in which a IIROC (2012), MAR Art. 12 (2)(c)(ii)
book fade) trader places an order away from the market
bid/ask on one side of the market and
subsequently submits increasing/decreasing
bids/asks for the same security on the other
side of the market without the intention of
the additional orders being executed to move
the market price in the direction of his first
order by creating a false impression of supply
and demand. Once the initial order is
executed, the trader cancels all other orders
and profits from the price reversal.

Table 6 Example for a taxonomy with two objects and two classes; the ‘‘X’’ or dimensions have been added/split/merged (objective
indicates the assignment (Nickerson et al., 2013) condition) (for a detailed set of possible combinations, see
Objects Dimension1 Nickerson et al., 2013). For developing the taxonomy, we
initially start with the four categories determined by the
Characteristic1 Characteristic2 classification of SEC LRs (see Table 3). Then, we integrate the
financial market manipulations found in Table 5. In each
Object1 X iteration, either the empirical-to-conceptual or the concep-
Object2 X tual-to-empirical approach can be followed to identify
characteristics and objects. If significant domain understand-
ing is available, the conceptual-to-empirical approach is
T1 ¼ fDimension1 ½Characteristic1 ; Characteristic2 g ð2Þ chosen (first, identification of new characteristics; thereafter,
determination of appropriate objects). In contrast, if signif-
Several iterations are accomplished to find possible icant data on the objects are available, the empirical-to-
dimensions that describe the manipulation techniques in conceptual approach is selected (first, identifying a set of
more detail until the ending conditions are met. The method objects; thereafter, identification of common characteristics).
ends when both subjective and objective conditions have In the following, we outline the different iterations of our
been met: (1) the taxonomy must be concise, robust, taxonomy development process (due to space constraints, the
comprehensive, extendible and explanatory (subjective con- identified objects are shown in the final result of the
ditions); and (2) in the last iteration, no (new) characteristics taxonomy development process).
 A taxonomy of financial market manipulations
Iteration 1
For the first iteration, the empirical-to-conceptual approach
is chosen; the objects of the first iteration are determined by
the four clusters that were found in the analysis of the SEC
LRs: financial instrument manipulation, accounting fraud,
investment fraud and insider trading.
Furthermore, the work of Allen and Gale (1992) is the
reference work in the manipulation literature and provides a
first categorization (Aggarwal and Wu, 2006; Kyle and
Viswanathan, 2008; Hillion and Suominen, 2004). According
to Allen and Gale (1992), financial market manipulations can
be classified by the means of the manipulative act. They
define three means of manipulation techniques: actions
which alter the actual or perceived value of the underlying
assets (action-based manipulation), information misusage by
disseminating false information or rumors (information-
based manipulation), and trading techniques to manipulate
the market (trade-based manipulation).
However, in today’s markets, with traders using low-
latency infrastructures, manipulation techniques also exist
that are based not on trades but only on orders because
traders make use of their superior latency to cancel orders
before they are executed (Biais and Woolley, 2011). There-
fore, we extend the categorization by Allen and Gale (1992)
with the category of order-based manipulation, leading to the
following taxonomy:
T1 ¼ fMeans ½Action based; Information based;
ð3Þ
Trade based; Order basedg
The ending conditions are not fully met. The subjective
ending conditions conciseness and comprehensiveness are
met, but the taxonomy is not yet robust in terms of covering
all manipulation techniques, its explanatory power is not
sufficient, and extendibility is not achieved. Furthermore,
new characteristics and dimensions were added. Therefore,
both objective and subjective ending conditions are not met,
and an additional iteration is required.
Iteration 2
The empirical-to-conceptual approach was chosen for the
second iteration. Specifically, more manipulation techniques
can be identified through the analysis provided in Table 5.
These techniques provide a closer view into the structure of
the previous objects because they split them into more
specific manipulation techniques.
Additionally, a new dimension can be added in this step by
analyzing the manipulator with respect to the employed
technique. Possible manipulators are issuers of a financial
asset, intermediaries and investors (Kyle and Viswanathan,
2008). Intermediaries are banks and brokers; however, they
can also act as investors if they trade on their own account.
Investors, therefore, are defined as market participants
trading on their own account, e.g., dealers, HFTs and other
ultimate recipients of rights in a security.
Not all ending conditions are met; specifically, the
taxonomy is not sufficiently explanatory because some
dimensions that could add explanatory power are not yet
included. Moreover, new objects and dimensions were added;
therefore, iteration 2 does not yet mark an ending point but
results in the following taxonomy:
M. Siering et al
T2 ¼ fMeans ½Action based; Information based; Trade based;
Order based; Manipulator ½Issuer; Intermediary; Investorg
ð4Þ
Iteration 3
For the next iteration, the conceptual-to-empirical approach
is chosen (Nickerson et al., 2013). No new objects can be
defined, because the list of objects in Table 5 is exhaustive
and reflects all of the techniques currently prosecuted by
regulators and discussed in academic studies. A new set of
characteristics can be found by describing the manipulation
target. Asset prices, volumes patterns and fundamental data
are often targeted by manipulators (Putnins, 2012). More-
over, the bid/ask spread, commissions and latency might be
possible targets of manipulation techniques. The knowledge
of manipulation targets can help officials and fraud detection
tools to prevent manipulation by closely monitoring the key
targets usually chosen by manipulators (Lenard and Alam,
2009). The previously found characteristics can be further
summarized under the dimension manipulation target:
T3 ¼ fMeans ½Action based; Information based; Trade based;
Order based; Manipulator ½Issuer; Intermediary; Investor;
Manipulation target ½Bid=Ask; Price; Volume;
Fundamental data; Commissions; Latencyg
ð5Þ
Most subjective ending conditions are met, but many of
the objects are not yet explicitly distinguishable through the
characteristics. Moreover, there was a new dimension added,
so an additional iteration is necessary.
Iteration 4
For this iteration, the conceptual-to-empirical approach is
chosen. We rely on the studies of Ledgerwood and Carpenter
(2012) and Aggarwal and Wu (2006), which provide
information about potential dimensions to be added.
The first characteristic to be added is whether a technique
is economically reasonable or based on uneconomic decisions
(Ledgerwood and Carpenter, 2012). A manipulation tech-
nique can include sell or purchase practices that are
specifically designed to lose money initially, compensated
by illicit profits later. This category can be of high importance
for the detection of market manipulation because uneco-
nomical trades should be rare in manipulation-free markets
and stand out when transactions are being monitored (Fama,
1970).
The second characteristic that can be added is whether a
manipulation technique has a specific direction to which the
target is manipulated (Aggarwal and Wu, 2006). This charac-
teristic is of high importance when distinguishing objects in a
reliable manner. For example, ‘‘Pump and Dump’’ and ‘‘Short
and Distort’’ cannot be discriminated through the taxonomy
from iteration 3. However, the initial action of ‘‘Pump and
Dump’’ (‘‘Short and Distort’’) has the goal of increasing
(decreasing) the price. Consequently, both can be distinguished.
In contrast, ‘‘Marking the Close’’ can lead to price manipulations
in both directions, depending upon the intention of the
 A taxonomy of financial market manipulations
manipulator. The subjective ending conditions of taxonomy
development are fully met, but in this iteration, new dimensions
were added; therefore, a new iteration is necessary:
T4 ¼ fMeans ½Action based; Information based; Trade based;
Order based; Manipulator ½Issuer; Intermediary; Investor;
Manipulation target ½Bid=Ask; Price; Volume;
Fundamental data; Commissions; Latency;
Economically reasonable ½Yes; No;
Specific target direction ½Up; Down; Both possible; Nog
ð6Þ
Iteration 5
For this iteration, the conceptual-to-empirical approach was
again chosen. The MAR states in article 12 (2)(a) that specific
times of the day can be used as an indicator for the
occurrence of certain manipulation strategies. Therefore, we
include the dimension specific point in time to distinguish
manipulation techniques that depend upon specific points in
time to be accomplished, e.g., ‘‘Marking the Close,’’ which is
only feasible at the end of the trading day. Additionally, the
MAR suggests that some fraudulent activities depend upon
the collusion of market participants. Therefore, we incorpo-
rate the dimension collusion in the taxonomy. ‘‘Benchmark
Manipulation’’ with and without official fixing is the prime
example of such a fraudulent activity. Furthermore, MAR
recital 38 points out that specific abusive strategies can be
performed by algorithmic traders and, in particular, HFTs
using low-latency infrastructure. Consequently, we added the
dimension low latency. Examples of such manipulative
strategies are ‘‘Quote Stuffing’’ and ‘‘Layering.’’ Due to three
new dimensions, an additional iteration is necessary. The
iteration results in the following taxonomy:
T5 ¼ fMeans ½Action based; Information based; Trade based;
Order based; Manipulator ½Issuer; Intermediary; Investor;
Manipulation target ½Bid=Ask; Price; Volume;
Fundamental data; Commissions; Latency;
Economically reasonable ½Yes; No;
Specific target direction ½Up; Down; Both possible; No;
Specific time ½Yes; No; Collusion½Yes; No;
Low latency ½Yes; Nog
ð7Þ
Iteration 6
An empirical-to-conceptual approach cannot be pursued in
this iteration because there are no further empirical findings
relevant for extending the taxonomy. Furthermore, no
important additional concepts remain; thus, nor can a
conceptual-to-empirical approach be applied. All ending
conditions, subjective and objective, are met, and the
taxonomy is provided by T5 (iteration 5). Therefore, the
final taxonomy of market manipulations in interaction with
the environment of the manipulated market is shown in
Table 7. Based on the methodology used by Nickerson et al.
(2013), we derived eight dimensions for our taxonomy of
M. Siering et al
market manipulations that can be used for the configuration
of an automated market surveillance system to detect
manipulative actions.
Evaluation of the taxonomy
Evaluation following Nickerson et al.
This section tests and discusses the developed taxonomy to
assess the contribution to the financial market manipulation
literature and the overall understanding of manipulations.
First, because the objective criterion is already fulfilled, the
subjective (qualitative) criteria are tested. Then, the practical
applicability and overall usefulness will be addressed.
Conciseness Nickerson et al. (2013) stated that an overly
broad taxonomy is a weak one. Because the cognitive capacity
of the decision maker is naturally limited, a taxonomy that
covers every characteristic and object possible is not concise
(Nickerson et al., 2013). The taxonomy developed above does
not suffer from this issue, because all characteristics are
documented in the literature, and most can be found in actual
LRs. Eight dimensions and 25 characteristics are also unlikely
to exceed the decision makers’ cognitive capacity (Carroll,
1993; Miller, 1956).
Robustness A robust taxonomy can clearly distinguish
objects from one another via characteristics and dimensions;
a small number of dimensions can hardly achieve this goal
(Nickerson et al., 2013). This taxonomy is able to distinguish
clearly two objects from one another; for example, the two
techniques ‘‘Pump and Dump’’ and ‘‘Scalping’’ are hard to
distinguish due to their similar appearance. The dimension
manipulator, however, helps to distinguish the two clearly
because ‘‘Pump and Dump’’ can be accomplished by
investors, whereas ‘‘Scalping’’ is conducted by intermediaries.
Comprehensiveness A taxonomy that is developed both
conceptually and empirically, such as this one, should cover all
dimensions and all objects possible to be viewed as compre-
hensible (Nickerson et al., 2013). An extensive literature review
and the use of EU legislation and market manipulations sued
by the SEC in the USA ensure that all objects (manipulation
techniques) currently considered relevant and all dimensions
that could possibly characterize those techniques are covered.
Extendibility A taxonomy should also be applicable in the
future, when new objects or dimensions extend the domain
for which the taxonomy was developed (Nickerson et al.,
2013). This criterion is hard to test because knowledge about
future forms of manipulation is only vague. However, the
taxonomy can be extended due to technological enhance-
ments because it is already capable of including manipulation
techniques associated with HFT, which represents a recent
major technological trend.
Explanatory power According to Nickerson et al. (2013), a
taxonomy should help to understand the domain without
describing every object in detail. The developed taxonomy
helps to understand the anatomy of market manipulations and
shows which manipulation techniques should be considered by
regulatory authorities. Nevertheless, the explanatory power is
not as strong for some objects as for others. For example, the
difference between ‘‘Matched Orders’’ and ‘‘Wash Sales’’ is only
identifiable based on whether collusion of market participants
is necessary.
Table 7 Taxonomy of financial market manipulations

Manipulation Manipulation Means of manipulation Manipulator Manipulation target

category technique
Action Information Trade Order Issuer Intermediary Investor Bid/ Price Volume Fundamental Commissions Latency
based based based based Ask
Accounting Fraudulent X X X
fraud financial
statements
Investment Benchmark X X X
fraud manipulation
(official fixing)
Benchmark X X X
manipulation
(no official
fixing)
Ponzi scheme X X X
Insider trading Insider trading X X X
Financial Pump and dump X X X
instrument Short and distort X X X
manipulation Cornering X X X
(squeezing)
A taxonomy of financial market manipulations

Advancing the X X X
bid
Reducing the ask X X X
Matched orders X X X
Painting the tape X X X
Wash sales X X X
M. Siering et al

Capping X X X
(pegging)
Marking the X X X
close/open
(banging the
close)
Front running X X X
Churning X X X
Scalping X X X
Spoofing X X X
Pinging X X X
Quote stuffing X X X
Ramping X X X
Layering (order X X X
book fade)
Table 7 continued

Manipulation Manipulation Specific Econ. Specific Collusion Low

category technique target reasonable time latency
direction
Up Down Both No Yes No Yes No Yes No Yes No
possible
Accounting fraud Fraudulent financial statements X X X X X
Investment fraud Benchmark manipulation (official X X X X X
fixing)
Benchmark manipulation (no X X X X X
official fixing)
Ponzi scheme X X X X X
Insider trading Insider trading X X X X X
Financial Pump and dump X X X X X
instrument Short and distort X X X X X
manipulation Cornering (squeezing) X X X X X
Advancing the bid X X X X X
Reducing the ask X X X X X
Matched orders X X X X X
Painting the tape X X X X X
A taxonomy of financial market manipulations

Wash sales X X X X X
Capping (pegging) X X X X X
Marking the close/open (banging X X X X X
the close)
Front running X X X X X
Churning X X X X X
M. Siering et al

Scalping X X X X X
Spoofing X X X X X
Pinging X X X X X
Quote stuffing X X X X X
Ramping X X X X X
Layering (order book fade) X X X X X
 A taxonomy of financial market manipulations
Application of the taxonomy in decision support system
development
To show the applicability of the taxonomy for DSS config-
uration, we focus on ‘‘Marking the Close’’ market manipu-
lations. We assume that the taxonomy is appropriate for
finding a proper DSS configuration if the specific configu-
ration (i.e., most importantly, the data sources considered)
leads to a successful system for fraud detection. Therefore,
the ‘‘Marking the Close’’ manipulation technique is suit-
able for investigating the appropriateness of the taxonomy
because different previous studies have proposed systems for
detecting manipulations in which the closing price of a
financial instrument is manipulated.
For marking the close, the means of market manipulation is
trade based. In other words, by means of specific trades, the
closing price is manipulated (dimension: manipulation
target). Consequently, an appropriate DSS should analyze
trading data including stock prices for manipulation detec-
tion. Because the closing price can be manipulated either
upwards or downwards, depending upon the specific goal of
the manipulator, both directions of manipulation are possi-
ble. Thus, a DSS should consider whether the suspicious
stock price consistently moves in one direction. In addition,
the manipulation focuses on a specific point in time, which
shows that in this case, closing prices should be considered by
DSSs. Nevertheless, no specific low-latency infrastructure is
necessary to pursue this manipulation technique, so the
utilization of such infrastructures cannot be considered a cue
for manipulation detection.
Furthermore, focusing on the manipulator performing the
manipulation, here the investor tries to perform marking the
close manipulations. This might be attempted, e.g., to avoid
margin calls, to support a flagging price or to affect the
valuation of a fund’s portfolio at the end of a specific relevant
period (e.g., quarter). Specifically, data identifying whether
an investor has an incentive because of his position in the
instrument might be helpful to identify related information.
Nevertheless, such data are only available for market
surveillance authorities and should therefore be considered
within their specific fraud detection systems. Concerning the
collusion dimension, note that no collusive behavior is
necessary for ‘‘Marking the Close’’; therefore, this dimension
should not be considered in a DSS.
Finally, the taxonomy also shows that the manipulative
behavior is not economically reasonable, i.e., in the case of a
marking the close manipulation, there should be other gains
for the investor to compensate him for performing this
costly activity, e.g., a higher valuation that is applied to
reveal a higher own portfolio performance of a fund
(window dressing). Therefore, data that allow crosschecking
positions of the respective investors should be used as
additional input.
Focusing on previous research to confirm whether the
proposed system configuration is actually implemented, we
consider the studies by Öğüt et al. (2009) and Kim and Sohn
(2012). Öğüt et al. (2009) consider average daily returns and
other daily trading statistics as input variables to detect
‘‘Marking the Close’’ by means of different machine learning
techniques. Additionally, Kim and Sohn (2012) consider
daily trading data encompassing closing prices to perform a
peer group analysis to detect suspicious behavior.
M. Siering et al
Both studies confirm the configuration as derived from the
taxonomy, and both studies consider trading data, including
stock prices. Furthermore, the studies consider a specific point
in time, i.e., the closing price. Because the closing price might
be manipulated in both directions depending upon the
manipulator’s intention, the direction is not directly consid-
ered, but a deviation from a benchmark is considered.
Because neither study has access to trading data, including
the market participant trading the financial instrument, they
are not able to consider whether the investor performs a
potential market manipulation. Finally, both studies could
also be improved if – in suspicious cases – the economic
reasonability of trades is assessed to identify investors who
have significant positions in the financial instrument under
investigation.
Consequently, the application of the taxonomy for DSS
configuration (including the comparison with previous
studies) shows that the taxonomy provides very useful
guidelines on the specific design of a DSS; it corresponds to
the design of existing systems and provides insights into
potential system enhancements.
Discussion
As shown by the evaluation of the taxonomy according to the
criteria proposed by Nickerson et al. (2013) and by the
application of the taxonomy for DSS configuration, the
proposed taxonomy is valid and very useful in the field of
classification and detection of market manipulations. It thus
helps to distinguish different forms of market manipulations
and to identify similarities between diverse types of manip-
ulative behavior. It also indicates clearly which information
an automated fraud detection system must incorporate. Due
to the comprehensive sources upon which the taxonomy is
built (cases actually prosecuted by a market surveillance
authority, cases reported in the literature and cases extracted
from financial market manipulation regulations), the taxon-
omy can be assumed to offer a broad overview on market
manipulations. Furthermore, the different dimensions
directly allow drawing conclusions on the specific configu-
ration of fraud detection systems. Consequently, the pro-
posed taxonomy can support market surveillance authorities
in detecting securities fraud to ensure fair and efficient
financial markets and a level playing field for market
participants, which is in the interests of all stakeholders,
particularly in the light of the financialization debate.
The distribution of the characteristics assigned to objects is
not even for all dimensions of the taxonomy. In particular,
the means of manipulation, the manipulator and the collusion
dimension place a great deal of weight on the characteristics
trade based, investor and no collusion necessary. This result
occurs because many financial instrument manipulations
have these characteristics in common. Nevertheless, these
dimensions are necessary because they represent key infor-
mation for regulators and automated fraud detection systems
trying to detect market manipulation and to differentiate
clearly between different manipulation techniques.
We are aware that fraudulent market participants might
also use the insights of the proposed taxonomy to circumvent
the risk of being detected by an automated fraud detection
system. Nevertheless, the taxonomy provides general insights
 A taxonomy of financial market manipulations
into DSS configuration that also apply in the case of changed
manipulator behavior. In this case, an appropriate DSS must
be recalibrated to detect the altered manipulative behavior.
The current study also has several limitations. The taxon-
omy provides implications for how to configure DSSs,
particularly in terms of the input data to be considered.
However, we are aware that the taxonomy does not itself
provide insights into how to process such potential input data
in a DSS. For instance, diverse machine learning algorithms
have been shown to lead to different performance figures. In
addition, from the specific point in time when a manipulation
is conducted, e.g., in the case of ‘‘Marking the Close,’’ further
aspects concerning the time passed for executing strategies
can hardly be considered within the taxonomy because
manipulators might chose different intervals for their manip-
ulations. Therefore, DSS developers considering the taxon-
omy should also perform different machine learning
experiments to find the best-performing configuration.
Finally, as already shown during the application of the
taxonomy in DSS development, not every dimension of a
market manipulation can be detected automatically. Never-
theless, the taxonomy provides a useful base to develop DSSs
that help to reduce information overload and to identify
cases that must be examined manually. Particularly against
the background of the increasing importance of financial
markets in the context of financialization, IS should consider
the taxonomy because domain knowledge for detecting
market manipulations can help to increase trust in and the
integrity of financial markets.
Conclusion
As long as financial markets have existed, fraudulent market
participants have tried to profit by manipulating markets and
deceiving others. Because market manipulations result in
substantial losses that must be borne by market participants,
understanding and detecting manipulations are of funda-
mental importance to establishing trust and market integrity –
particularly against the background of financialization and the
role of markets in financialization economies. Nevertheless,
until now, different terms have been used even for the same
market manipulation techniques, and no unified terminology
prevails, which also hampers proper fraud detection.
Within this study, we follow a multi-method approach to
develop a taxonomy of financial market manipulations. In
particular, we consider officially prosecuted current market
manipulations by performing a cluster analysis on SEC LRs.
We also perform a literature review and a regulatory review to
identify market manipulations and their characteristics. On
that basis, we develop a novel taxonomy that covers existing
market manipulation techniques, including those based on
HFT. Applying eight dimensions, the taxonomy helps to
distinguish clearly between different fraudulent activities and
shows which information must be considered by automated
fraud detection systems. Based on the ‘‘Marking the Close’’
manipulation, we successfully evaluate the taxonomy and show
how to employ the taxonomy to build a fraud detection system.
Guided by this taxonomy, researchers and practitioners
developing DSSs can find an appropriate system configuration.
We provide a solid understanding of the different manip-
ulation techniques and their characteristics. We thereby
extend previous studies and contribute to the literature by
M. Siering et al
specifically considering the manipulation techniques actually
performed and prosecuted and by specifically considering
manipulation techniques caused by novel technological
developments such as HFT. Therefore, our study also
contributes to the call by Ngai et al. (2011), who find that
there is a lack of research in the field of securities fraud
detection; the study shows how the taxonomy can be used as
a foundation for DSSs helping to detect fraud.
Our results are also relevant for practitioners, regulators
and financial market participants. First, our study provides an
overview of current financial market manipulations and their
characteristics. This classification can provide an important
means for increasing the financial literacy of both institutional
and private investors by showing them potential market
manipulations and for generating awareness concerning
suspicious market behavior. Based on this research, financial
institutions and market surveillance authorities can develop
DSSs that help to identify market manipulations. Conse-
quently, these organizations are enabled to enforce recent
market regulations to ensure fair and efficient financial
markets and thus to alleviate the negative effect of fraud on
the economy, particularly in the light of financialization. Our
results are also of relevance for software vendors because the
study provides indications on how to configure appropriate
DSSs to detect financial market manipulations. As also
identified during the taxonomy development, HFT is accom-
panied by novel manipulation techniques. Because our paper
shows which type of DSS configuration might help to identify
related fraudulent activities, future research might build upon
the taxonomy to develop specific deception detection systems.
Given the massive increase in regulation of financial
markets and the banking industry as well as the need to
provide efficient and low-cost technological support and
automation in this field, a novel future research stream ‘‘Reg-
Tech’’ will be an interesting and highly relevant topic for
further research at the intersection of financialization and IS.
Reg-Tech encompasses ‘‘the adoption of new technologies to
facilitate the delivery of regulatory requirements’’ (Financial
Conduct Authority (FCA), 2015). Our work is one step in
that direction; however, additional future research in the
fields of, for example, artificial intelligence, big data analysis
and text mining techniques might provide very useful input
for investment firms, regulators, digital solution providers
and consultancy houses to fulfill ever-growing regulatory
requirements in banking and financial markets.
References
Aalbers, M. (2016). Corporate financialization. In D. Richardson, N. Castree, M.
F. Goodchild, A. Kobayashi, W. Liu, and R. A. Marston (Eds.), International
Encyclopedia of Geography: People, the Earth, Environment and Technology.
Oxford: Wiley.
Abbasi, A., Zhang, Z., Zimbra, D., Chen, H., Nunamaker, Jr, and Jay, F. (2010).
Detecting Fake Websites: The Contribution of Statistical Learning Theory. MIS
Quarterly, 34(3), 435–461.
Abrantes-Metz, R. M., Kraten, M., Metz, A. D., and Seow, G. S. (2012). Libor
Manipulation? Journal of Banking & Finance, 36(1), 136–150.
Aggarwal, R. K., and Wu, G. (2006). Stock Market Manipulations. Journal of
Business, 79(4), 1915–1953.
Akhigbe, A., Madura, J., and Martin, A. (2005). Accounting Contagion: The Case
of Enron. Journal of Economics and Finance, 29(2), 187–202.
Aliguliyev, R. M. (2009). Performance Evaluation of Density-Based Clustering
Methods. Information Sciences, 179(20), 3583–3602.
 A taxonomy of financial market manipulations
Allen, F., and Gale, D. (1992). Stock-Price Manipulation. Review of Financial
Studies, 5(3), 503–529.
Allen, F., Litov, L., and Mei, J. (2006). Large Investors, Price Manipulation, and
Limits to Arbitrage: An Anatomy of Market Corners. Review of Finance, 10(4),
645–693.
Allen, S., and Ramanan, R. (1995). Insider Trading, Earnings Changes, and Stock
Prices. Management Science, 41(4), 653–668.
Arshadi, N. (1998). Insider Trading Liability and Enforcement Strategy. Financial
Management, 27(2), 70–84.
Bailey, K. D. (1984). A Three-Level Measurement Model. Quality & Quantity,
18(3), 225–245.
Baker, N. (2005). Fraud and Artificial Intelligence. Internal Auditor, 62(1), 29–32.
Bartels, K. C. (2000). Click Here to Buy the Next Microsoft: The Penny Stock
Rules, Online Microcap Fraud, and the Unwary Investor. Indiana Law Journal,
75(1), 353–378.
Biais, B. and Woolley, P. (2011). High frequency trading [WWW document]
http://www.eifr.eu/files/file2220879.pdf (accessed 24th February 2016).
Bonner, S. E., Palmrose, Z.-V., and Young, S. M. (1998). Fraud Type and
Auditor Litigation: An Analysis of SEC Accounting and Auditing Enforcement
Releases. The Accounting Review, 73(4), 503–532.
Bouwman, M. J. (1983). Human Diagnostic Reasoning by Computer: An
Illustration from Financial Analysis. Management Science, 29(6), 653–672.
Carhart, M. M., Kaniel, R., Musto, D. K., and Reed, A. V. (2002). Leaning for the
Tape: Evidence of Gaming Behavior in Equity Mutual Funds. The Journal of
Finance, 57(2), 661–693.
Carroll, J. B. (1993). Human Cognitive Abilities: A Survey of Factor-Analytic
Studies. Cambridge: Cambridge University Press.
Carroll, B. (2006). How to Prevent Investment Adviser Fraud. Journal of
Accountancy, 201(1), 40–43.
Cataldo, A. J., and Killough, L. N. (2003). Market Makers’ Methods of Stock
Manipulation. Management Accounting Quarterly, 4(4), 10–13.
Chaturvedula, C., Bang, N. P., Rastogi, N., and Kumar, S. (2015). Price
Manipulation, Front Running and Bulk Trades: Evidence from India. Emerging
Markets Review, 23, 26–45.
Coffee, J. C. (2005). A Theory of Corporate Scandals: Why the USA and Europe
Differ. Oxford Review of Economic Policy, 21(2), 198–211.
Comerton-Forde, C., and Putnins, T. J. (2011). Measuring Closing Price
Manipulation. Journal of Financial Intermediation, 20(2), 135–158.
Cooper, D. J., and Donaldson, R. G. (1998). A Strategic Analysis of Corners and
Squeezes. The Journal of Financial and Quantitative Analysis, 33(1), 117–137.
Cumming, D., and Johan, S. (2008). Global Market Surveillance. American Law
and Economics Review, 10(2), 454–506.
Cumming, D., Zhan, F. and Aitken, M. (2012). High Frequency Trading and
End-of-Day Manipulation [WWW document] https://www.legacy.wlu.ca/
documents/54105/Cumming-Zhan-Aitken-15012013_2.pdf (accessed 24th
February 2016).
Davis, H. A. (2007). Summary of Selected FINRA Regulatory Notices. Journal of
Investment Compliance, 8(4), 60–67.
Denney, A. S., and Tewksbury, R. (2013). How to Write a Literature Review.
Journal of Criminal Justice Education, 24(2), 218–234.
Easley, D., Prado, De, Lopez, Marcos M., and O’Hara, M. (2011). The
Microstructure of the ‘‘Flash Crash’’: Flow Toxicity, Liquidity Crashes,
and the Probability of Informed Trading. Journal of Portfolio Management,
37(2), 118.
European Parliament and Council (2014). Market Abuse Regulation No
596/2014 [WWW document] http://eur-lex.europa.eu/legal-content/EN/TXT/
?uri=celex%3A32014R0596 (accessed 24th February 2016).
European Parliament and Council (2016). Benchmark Regulation 2016/1011
[WWW document] http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=
CELEX:32016R1011&from=EN (accessed 4th August 2016).
Fama, E. F. (1970). Efficient Capital Markets: A Review of Theory and Empirical
Work. The Journal of Finance, 25(2), 383–417.
Fatudimu, I., Musa, A., Ayo, C., and Sofoluwe, A. B. (2008). Knowledge
Discovery in Online Repositories: A Text Mining Approach. European Journal
of Scientific Research, 22(2), 241–250.
Fayyad, U., Piatetsky-Shapiro, G., and Smyth, P. (1996). From Data Mining to
Knowledge Discovery in Databases. AI Magazine, 17(3), 37–54.
Financial Conduct Authority (FCA) (2014). FCA Final Notice: Reference
Number 124491 [WWW document] http://www.fca.org.uk/your-fca/
documents/final-notices/2014/jpmorgan-chase-bank (accessed 25th August
2016).
M. Siering et al
Financial Conduct Authority (FCA) (2015). Call for Input: Supporting the
Development and Adoption of RegTech [WWW document] https://www.fca.
org.uk/publication/call-for-input/regtech-call-for-input.pdf (accessed 12th
September 2016).
Financial Services Authority (FSA) (2012). FSA Final Notice: Reference
Number 122702 [WWW document] http://www.fsa.gov.uk/static/pubs/final/
barclays-jun12.pdf (accessed 25th August 2016).
Fischel, D. R., and Ross, D. J. (1991). Should the Law Prohibit ‘‘Manipulation’’ In
Financial Markets? Harvard Law Review, 105(2), 503–553.
Fouquau, J., and Spieser, P. K. (2015). Statistical Evidence About LIBOR
Manipulation: A ‘‘Sherlock Holmes’’ Investigation. Journal of Banking &
Finance, 50, 632–643.
Frank, J. (1994). Artificial Intelligence and Intrusion Detection: Current and
Future Directions, in 17th National Computer Security Conference; Baltimore,
USA, 1994, pp. 1–12.
Gerety, M., and Lehn, K. (1997). The Causes and Consequences of Accounting
Fraud. Managerial and Decision Economics, 18(7–8), 587–599.
Golmohammadi, K., Zaiane, O.R. and Diaz, D. (2014). Detecting Stock Market
Manipulation Using Supervised Learning Algorithms, in International Con-
ference on Data Science and Advanced Analytics (DSAA); Shanghai, China,
2014, pp. 435–441.
Gomber, P., Sagade, S., Theissen, E., Weber, M.C. and Westheide, C. (2016).
Competition Between Equity Markets: A Review of the Consolidation Versus
Fragmentation Debate, Journal of Economic Surveys (forthcoming).
Hanke, M., and Hauser, F. (2008). On the Effects of Stock Spam e-Mails. Journal
of Financial Markets, 11(1), 57–83.
Hazen, T. L. (2010). Are Existing Stock Broker Standards Sufficient? Principles,
Rules and Fiduciary Duties. Columbia Business Law Review, 2010(3), 710–761.
Hellwig, M. F. (1980). On the Aggregation of Information in Competitive
Markets. Journal of Economic Theory, 22(3), 477–498.
Hillion, P., and Suominen, M. (2004). The Manipulation of Closing Prices.
Journal of Financial Markets, 7(4), 351–375.
Humpherys, S. L., Moffitt, K. C., Burns, M. B., Burgoon, J. K., and Felix, W. F.
(2011). Identification of Fraudulent Financial Statements Using Linguistic
Credibility Analysis. Decision Support Systems, 50(3), 585–594.
IIROC (2012). Proposed Guidance on Certain Manipulative and Deceptive
Trading Practices [WWW document] http://www.iiroc.ca/Documents/2012/
f62c746a-b5c9-448a-b57f-f1c04c88de14_en.pdf (accessed 19th February 2016).
Ionescu, L. (2010). Madoff’s Fraudulent Financial Scheme, His Decades-Long
Swindle, and the Failure of Operational Risk Management. Economics,
Management, and Financial Markets, 5(3), 239–244.
Jarrow, R. A. (1992). Market Manipulation, Bubbles, Corners, and Short
Squeezes. Journal of Financial and Quantitative Analysis, 27(03), 311–336.
Jones, K. S. (1972). A Statistical Interpretation of Term Specificity and Its
Application in Retrieval. Journal of Documentation, 28(1), 11–21.
Jones, M. J. (2011). Creative Accounting, Fraud And International Accounting
Scandals. Hoboken: Wiley.
Karz, G., and Wagner, W. H. (2006). Should I Fire My Trader or Pay Him A
Million? The Journal of Trading, 1(4), 85–89.
Kim, Y., and Sohn, S. Y. (2012). Stock Fraud Detection Using Peer Group
Analysis. Expert Systems with Applications, 39(10), 8986–8992.
Kirkos, E., Spathis, C., and Manolopoulos, Y. (2007). Data Mining Techniques
for the Detection of Fraudulent Financial Statements. Expert Systems with
Applications, 32(4), 995–1003.
Klein, L. R., Dalko, V., and Wang, M. (2012). Regulating Competition in Stock
Markets: Antitrust Measures to Promote Fairness and Transparency Through
Investor Protection and Crisis Prevention. Hoboken: Wiley.
Kumar, P., and Langberg, N. (2009). Corporate Fraud and Investment
Distortions in Efficient Capital Markets. The Rand Journal of Economics, 40(1),
144–172.
Kyle, A. S., and Viswanathan, S. (2008). How to Define Illegal Price
Manipulation. American Economic Review, 98(2), 274–279.
Lagoarde-Segot, T. (2016). Financialization: Towards a New Research Agenda,
International Review of Financial Analysis. doi:10.1016/j.irfa.2016.03.007.
Ledgerwood, S. D., and Carpenter, P. (2012). A Framework for the Analysis of
Market Manipulation. The Review of Law & Economics, 8(1), 253–295.
Lee, E. J., Eom, K. S., and Park, K. S. (2013). Microstructure-Based Manipu-
lation: Strategic Behavior and Performance of Spoofing Traders. Journal of
Financial Markets, 16(2), 227–252.
Leinweber, D. J., and Madhavan, A. N. (2001). Three Hundred Years of Stock
Market Manipulations. The Journal of Investing, 10(2), 7–16.
 A taxonomy of financial market manipulations
Lenard, M. J., and Alam, P. (2009). An Historical Perspective on Fraud
Detection: From Bankruptcy Models to Most Effective Indicators of Fraud in
Recent Incidents. Journal of Forensic & Investigative Accounting, 1(1), 1–27.
Levy, Y., and Ellis, T. J. (2006). A Systems Approach to Conduct an Effective
Literature Review in Support of Information Systems Research. Informing
Science: International Journal of an Emerging Transdiscipline, 9(1), 181–212.
Lomnicka, E. (2001). Preventing and Controlling the Manipulation of Financial
Markets: Towards a Definition of ‘‘Market Manipulation’’. Journal of Financial
Crime, 8(4), 297–304.
Losiewicz, P., Oard, D. W., and Kostoff, R. N. (2000). Textual Data Mining to
Support Science and Technology Management. Journal of Intelligent Informa-
tion Systems, 15(2), 99–119.
LR 17346 (2002). Litigation Release No. 17346: Securities and Exchange
Commission v. Patrick O. Wheeler, Steven S. Gallers, and Robert L. Carberry,
Case No. 02-60131-CIV-GRAHAM (S.D. Fla.) [WWW document] http://www.
sec.gov/litigation/litreleases/lr17346.htm (accessed 19th February 2016).
LR 17645 (2002). Litigation Release No. 17645: Securities and Exchange
Commission v. Michael A. Ofstedahl, et al., United States District Court for the
Northern District of California, Civil Action No. C-02-3685 RS. [WWW
document] http://www.sec.gov/litigation/litreleases/lr17645.htm (accessed 19th
February 2016).
LR 21423 (2010). Litigation Release No. 21423: Securities and Exchange
Commission v. Frank J. Custable, Jr., et al., Civil Action No. 03-CV-2182 (N.D.
Ill.) [WWW document] http://www.sec.gov/litigation/litreleases/2010/lr21423.
htm (accessed 19th February 2016).
LR 22545 (2012). Litigation Release No. 22545: SEC v. Berton M. Hochfeld et al.,
Civil Action No. 12-CV-8202 (S.D.N.Y.) [WWW document] http://www.sec.
gov/litigation/litreleases/2012/lr22545.htm (accessed 19th February 2016).
MacQueen, J. (1967). Some Methods for Classification and Analysis of
Multivariate Observations, in Fifth Berkeley Symposium on Mathematical
Statistics and Probability (Berkeley, USA, 1967), pp. 281–297.
Miller, G. A. (1956). The Magical Number Seven, Plus or Minus Two: Some
Limits on Our Capacity for Processing Information. Psychological Review, 63(2),
343–352.
Miller, J. G., and Roth, A. V. (1994). A Taxonomy of Manufacturing Strategies.
Management Science, 40(3), 285–304.
Møllgaard, H. P. (1997). A Squeezer Round the Corner? Self-Regulation and
Forward Markets. The Economic Journal, 107(440), 104–112.
Ngai, E., Hu, Y., Wong, Y., Chen, Y., and Sun, X. (2011). The Application of
Data Mining Techniques in Financial Fraud Detection: A Classification
Framework and an Academic Review of Literature. Decision Support Systems,
50(3), 559–569.
Nickerson, R. C., Varshney, U., and Muntermann, J. (2013). A Method for
Taxonomy Development and Its Application in Information Systems. European
Journal of Information Systems, 22(3), 336–359.
Öğüt, H., Doğanay, M. M., and Aktaş, R. (2009). Detecting Stock-Price
Manipulation in an Emerging Market: The Case of Turkey. Expert Systems with
Applications, 36(9), 11944–11949.
Pagano, M., and Immordino, G. (2012). Corporate Fraud, Governance, and
Auditing. Review of Corporate Finance Studies, 1(1), 109–133.
Pelleg, D. and Moore, A. (2000). X-Means: Extending K-Means with Efficient
Estimation of the Number of Clusters, in 17th International Conference on
Machine Learning; Stanford, USA, 2000, pp. 727–734.
Porter, M. F. (2006). An Algorithm for Suffix Stripping. Electronic Library and
Information Systems, 40(3), 211–218.
Pozza, C. L., Jr., Cox, T. R., and Morad, R. J. (2009). Review of Recent Investor
Issues in the Madoff, Standford and Forte Ponzi Scheme Cases. Journal of
Business and Securities Law, 10, 113–132.
Putnins, T. J. (2012). Market Manipulation: A Survey. Journal of Economic
Surveys, 26(5), 952–967.
Rapoport, N. B. (2012). Black Swans, Ostriches, and Ponzi Schemes. Golden Gate
University Law Review, 42, 627–661.
Rehman, Z., Anwar, W., Bajwa, U. I., Xuan, W., Chaoying, Z., and Patterson, R.
L. (2013). Morpheme Matching Based Text Tokenization for a Scarce
Resourced Language. PLoS ONE, 8(8), e68178.
Roddenberry, S., and Bacon, F. (2011). Insider Trading and Market Efficiency:
Do Insiders Buy Low and Sell High? Journal of Finance & Accountancy, 8, 1–15.
Rokach, L. and Maimon, O. (2005). Clustering Methods, in O. Maimon and L.
Rokach, (eds.), Data Mining and Knowledge Discovery Handbook, New York:
Springer US, pp. 321–352.
M. Siering et al
Rowley, J., and Slack, F. (2004). Conducting a Literature Review. Management
Research News, 27(6), 31–39.
Sabherwal, S., Sarkar, S. K., and Zhang, Y. (2011). Do Internet Stock Message Boards
Influence Trading? Evidence from Heavily Discussed Stocks with no Fundamental
News. Journal of Business Finance & Accounting, 38(9–10), 1209–1237.
Scopino, G. (2015). The (Questionable) Legality Of High-Speed Pinging and
Front Running in the Futures Market. Connecticut Law Review, 47(3),
607–697.
Shapiro, S., Kinkela, K., and Harris, P. (2012). Churning and Suitability of
Investments: A Financial Industry Regulatory Authority Arbitration Case
Study. Review of Business & Finance Case Studies, 3(1), 61–67.
Shleifer, A., and Vishny, R. W. (1997). A Survey of Corporate Governance. The
Journal of Finance, 52(2), 737–783.
Tabrizi, S. A., Shakery, A., Asadpour, M., Abbasi, M., and Tavallaie, M. A.
(2013). Personalized PageRank Clustering: A Graph Clustering Algorithm
Based on Random Walks. Physica A: Statistical Mechanics and its Applications,
392(22), 5772–5785.
Thel, S. (1993). 850,000 in Six Minutes-The Mechanics of Securities Manipu-
lation. Cornell Law Review, 79(2), 219–298.
Tripp, M. J. W. (1963). Securities Regulation: Stock Scalping by the Investment
Adviser: Fraud or Legitimate Business Practice? California Law Review, 51(1),
232–245.
U.S. Commodity Futures Trading Commission (CFTC) (2016). Education
Center: CFTC Glossary [WWW document] http://www.cftc.gov/
consumerprotection/educationcenter/cftcglossary/glossary_b (accessed 4th
August 2016).
Vaughan, L. (2016). Broken Benchmarks: Six Years of Probes into Financial
Fiddling [WWW document] https://www.bloomberg.com/quicktake/broken-
benchmarks (accessed 25th August 2016).
Vila, J.-L. (1989). Simple Games of Market Manipulation. Economics Letters,
29(1), 21–26.
Webster, J., and Watson, R. T. (2002). Analyzing the Past to Prepare for the
Future: Writing a Literature Review. Management Information Systems Quar-
terly, 26(2), 3.
Wheeler, R., and Aitken, S. (2000). Multiple Algorithms for Fraud Detection.
Knowledge-Based Systems, 13(2), 93–99.
Willett, P. (1988). Recent Trends in Hierarchic Document Clustering: A Critical
Review. Information Processing and Management, 24(5), 577–597.
Willett, P. (2006). The Porter Stemming Algorithm: Then and Now. Electronic
Library and Information Systems, 40(3), 219–223.
Ye, J. (2011). Cosine Similarity Measures for Intuitionistic Fuzzy Sets and Their
Applications. Mathematical and Computer Modelling, 53(1), 91–97.
Zahedi, F. M., Abbasi, A., and Chen, Yan. (2015). Fake-Website Detection Tools:
Identifying Elements that Promote Individuals’ Use and Enhance Their
Performance. Journal of the Association for Information Systems, 16(6), 448–484.
Zhou, L., Burgoon, J. K., Twitchell, D. P., Qin, T., Nunamaker, Jr, and Jay, F.
(2004). A Comparison of Classification Methods for Predicting Deception in
Computer-Mediated Communication. Journal of Management Information
Systems, 20(4), 139–166.
About the Authors
Michael Siering is a Postdoctoral Research Associate at
Goethe University Frankfurt and works as a Business
Consultant in the field of risk management. His research
focuses on decision support systems in electronic markets,
with a focus on the analysis of user generated content. His
work has been published in Decision Support Systems and
the Journal of Management Information Systems.
Benjamin Clapham (clapham@wiwi.uni-frankfurt.de) is a
Graduate Researcher at Goethe University Frankfurt. He
received his M.Sc. degree in Management with majors in
Finance and Accounting from Goethe University. His
 A taxonomy of financial market manipulations
research interests include market microstructure, algorithmic
trading, sentiment analysis, and financial market
manipulations.
Oliver Engel (oliengel@gmx.de) holds a M.Sc. degree in
Management from Goethe University Frankfurt, with a
specialization on e-finance. His research focuses on text
mining and financial market manipulations.
M. Siering et al
Peter Gomber (gomber@wiwi.uni-frankfurt.de) holds the
Chair of e-Finance at the Faculty of Economics and Business
Administration, University of Frankfurt. His academic work
focuses on market microstructure and auction theory,
institutional trading, innovative concepts/technologies for
electronic trading and post trading systems and information
systems in Finance.