Understanding the different types of financial scams and frauds targeting consumers can help public authorities more effectively target their efforts to combat financial scams and frauds. This chapter discusses the importance of collecting data on financial scams and frauds and sets forth a typology to help public authorities systematically collect, classify and monitor data, identify patterns or high-risk areas, and track trends.
Protecting Consumers from Financial Scams and Frauds
Abstract
To effectively combat financial scams and frauds, public authorities should understand the different types of scams and frauds targeting financial consumers. To fully understand the broader environment and the impact of financial scams and frauds, public authorities should collect, or have access to, data on the types, incidence and severity (i.e. amount of financial loss) of financial scams and frauds.
Furthermore, if such data is organised around a typology or classification system, public authorities can engage in more systematic and consistent recording, monitoring and analysis of different types of financial scams and frauds. Using such a classification system enables policymakers and public authorities to identify patterns or high-risk areas, track trends and inform decisions about allocation of resources to combatting the most common types of scams and frauds or those that result in the largest amount of financial loss to consumers.
Capturing different types of financial scams and frauds can also help consumers more accurately report what they experienced and inform consumer awareness efforts and communication campaigns with the public. If available, information about fraud victims’ socio-demographic characteristics or circumstances can help public authorities target their consumer outreach efforts and better understand which characteristics or circumstances may heighten vulnerability to detriment arising from financial scams and frauds. Lastly, data on financial scams and frauds can help public authorities evaluate the impact of their policies and interventions.
Forty-seven respondents indicated that the data on financial scams and frauds in their jurisdiction is reported using some type of typology or classification system. Five respondents indicated that a typology or classification system of financial scams and frauds was under development.
Therefore, half of the respondents do not currently have a typology or classification system for financial scams and frauds to engage in more systematic and consistent recording, monitoring and analysis of fraud data.
Regarding those respondents that already use a typology or classification system, Figure 3.1 shows the criteria used to classify data in existing typologies. The three most common criteria used to classify data on financial scams and frauds are:
the type of financial product or service affected
the conduct or tactic of the financial fraudster or scammer
and the contact method used to engage with the victim.
Additionally, two respondents reported keeping data on prevented cases versus executed cases, or the number of fraud attempts that were successfully stopped versus the number of attempts that were successfully carried out. Seven respondents also indicated that a market or industry standard was used to classify financial scams and frauds.
Source: OECD Questionnaire on Protecting Consumers from Financial Scams and Frauds (2025)
These criteria are operationalised in a variety of ways among respondents with a typology or classification system. For example, Bank Indonesia classifies fraud and scam reports based on the modus operandi and the type of financial product or service affected. In Colombia, the Superintendencia Financiera de Colombia (Financial Superintendency of Colombia, SFC) reports using twenty distinct fraud modalities,1 while the National Policy Agency in Japan applies a 10-category scheme for frauds targeting financial consumers, alongside separately tracking social-media-based investment and romance scams. In Canada for instance, typologies are based on the initial offer or solicitation received by the victim (e.g. a random text message used to develop a relationship; a targeted email received by someone posing as an employer asking for help; a call or instant message from a family member who is in trouble). The Canadian Anti-Fraud Centre then uses broader categories such as merchandise fraud, service fraud or emergency fraud and applies tags to specific variations of operational interest.
The National Bank of Ukraine, for example, receives data from payment service providers on losses from fraudulent transactions, which includes information on the location where the fraudulent transaction took place (e.g. ATM, POS terminal, internet, payment app, online banking site), the type of fraud (forged electronic payment instruments, lost/stolen instruments, compromised instruments, social engineering or other type of fraud), the entity that incurred the loss, the name of the payment system where the fraudulent transaction occurred, information about the network’s owner and the electronic payment instrument’s issuer, and the territory where the fraudulent transaction took place (within or outside Ukraine). In South Africa, the Financial Sector Conduct Authority initiated a Digital Fraud Project, with participation from across the financial sector and other relevant industries, and through this workstream identified 73 typologies of fraud.
Many respondents within the EU, including the Banca d’Italia, use a classification system based on the industry-led taxonomy developed by the Euro Banking Association (EBA Association), called the “the EBA Fraud Taxonomy”.2 This includes data on the type of instrument affected (money transfers, credit cards, debit cards, e-money cards), the channel where the fraud is executed, whether it was a national or international payment, whether it was a card-present or card-not-present transaction, the initiation channel, and whether the payment used strong customer authentication or not. This classification framework is limited to payment services fraud and therefore does not include other fraud types such as investment scams.
In France for example, the Autorité des Marchés Financiers (AMF) classifies reported incidents through keywords; currently, these keywords include crypto assets, fake green investments, fake savings accounts, Forex, precious stones/gold, fake listed securities, trading bots, fake real estate funds, and wine investments. When there is a new type of scam, the AMF can add new keywords. Similarly, at the Austrian Financial Market Authority, reported incidents of investment fraud are categorised as “traditional”, “crypto” and “identity theft of the identity of authorities”, along with information on the initial contact with citizens and the amount of financial loss.
In response to the Questionnaire, some respondents also provided explanations or definitions of common types of financial scams and frauds. A glossary is provided in Annex B.
While respondents acknowledged the value of a classification system or typology of financial scams, they consistently noted a variety of challenges in developing and using such a system. For example, Bank Indonesia noted that scammers and fraudsters continuously adapt and innovate their methods, which makes creating an exhaustive list of types of scams and frauds targeting financial consumers very challenging. Additionally, categories of financial scams and frauds may not be mutually exclusive, and consumers might be unsure which category(ies) apply when reporting incidents. A lack of a centralised classification system can be confusing for consumers, financial services providers, policymakers, regulators and supervisors alike. The Superintendencia de Banca, Seguros y AFP (Superintendency of Banking, Insurance and Private Pension Funds, SBS) in Peru also noted in its response that despite supervised institutions maintaining detailed complaint databases and periodic regulatory reports that support the classification and monitoring of fraud-related incidents across products, channels, and transaction types, there remain certain limitations in further enriching the analysis of fraudulent events. For instance, in practice, institutions often prioritise the immediate takedown of malicious websites and channels upon detection. While this rapid response helps mitigate risks to consumers, it limits the time and information available to analyse the full content of such fraudulent schemes, which could otherwise contribute to a more comprehensive understanding of fraud typologies.
In some jurisdictions, responsibility for regulating and supervising the financial sector is shared across different public authorities, resulting in fragmented data on financial scams and frauds. Such fragmentation across different public authorities can lead to concerns over data quality, duplication, comparability and gaps. Indeed, addressing financial scams and frauds is not limited to the financial sector and financial services regulators and supervisors; law enforcement and other public authorities involved in the anti-fraud ecosystem could use and benefit from a shared typology. In other words, without a common typology or classification system, data sharing across public authorities can prove challenging. Furthermore, engaging in cross-border information sharing can also prove challenging when jurisdictions use different systems and typologies.
Even with an agreed typology, respondents shared how collaboration could be further enhanced with an integrated national data platform. Many respondents, including for example the Australian Securities and Investments Commission (ASIC), the Brunei Darussalam Central Bank, the Central Bank of Ireland and the United Kingdom Financial Conduct Authority (FCA) shared that both a classification system and a centralised data platform would help foster co-operation across agencies, thereby promoting early detection and co-ordinated responses in jurisdictions with multiple financial sector regulators and supervisors.
This section puts forth ten key dimensions to classify financial scams and frauds, which together form the proposed typology. These dimensions incorporate and expand upon the criteria shared in the previous section. Figure 3.2 shows the ten key dimensions that together form the proposed typology to classify financial scams and frauds targeting consumers.
Ten key dimensions of the proposed typology to classify financial scams and frauds targeting consumers
Within each dimension, detailed incident-related information can be collected and utilised to classify and catalogue cases:
3.3.1. Target
The first dimension is the target of the financial scam or fraud. Basic socio-demographic information about the individuals targeted by the financial scam or fraud, such as their sex and/or gender,3 age and employment status would be useful for authorities to further understand whether and how certain groups of consumers may be at risk of different kinds of fraud. Therefore, dedicated reporting channels should provide the option for victims to share basic socio-demographic information about themselves, including characteristics or circumstances that may heighten their vulnerability to detriment arising from financial scams and frauds. To the extent that such details are available to financial services providers, given data privacy laws, aggregated information about the victims is useful for institutions to identify trends and tailor communication and awareness efforts.
3.3.2. Contact medium
The second dimension is the contact medium or form, i.e. how the fraudster or scammer first contacted victims. Sub-categories in this dimension include contacting the victim by phone, fax, SMS or text message, e-mail, postal mail, social media message or through online advertisements. The advantage of having this information is to understand the communication channels whereby victims are exposed to contact from fraudsters and scammers. Depending on the channel, this could provide further evidence of the importance of cross-sectoral collaboration.
3.3.3. Scheme or modus operandi
This dimension captures the method(s) fraudsters and scammers employ to trick their victims. This dimension may include multiple sub-categories including investment fraud, retail fraud, social engineering, predatory financial fraud, identity fraud, and banking fraud. Further descriptive categories may be used in this dimension to catalogue new and emerging fraud types. This dimension provides insights into the successful ways fraudsters and scammers are able to defraud victims.
3.3.4. Financial product or service targeted
As the title states, this dimension captures which financial product(s) or service(s) were targeted by the fraudster or scammer. This includes deposit accounts, debit or payment cards, credit cards, other credit accounts, mobile banking applications, electronic or digital wallets, insurance products, pensions or investments. This information helps authorities understand which product(s) or service(s) are more likely to expose consumers to risk, and/or which products and services are more easily manipulated by fraudsters and scammers.
3.3.5. Method of payment or investment of funds
This dimension captures the way that money was transferred from the victim to the fraudster or scammer. This includes cash, gift cards, crypto assets, debit cards, electronic or wire transfers and mobile payments. This dimension can also be further sub-categorised by the channel through which the fraud was executed, such as an ATM, POS terminal or online application.
3.3.6. Detection method
This dimension captures how the financial scam or fraud was initially detected, i.e. whether the consumer or the financial service provider first suspected fraud. In the case where the consumer first detected (or suspected) a financial scam or fraud, this dimension provides insight into levels of consumer awareness or knowledge about how to recognise financial scams and frauds. In the case where the financial service provider first detected a financial scam or fraud, this provides evidence about the deployment of effective transaction monitoring mechanisms and their ability to identify suspicious transactions.
3.3.7. Recovery status of funds
This dimension captures whether funds were fully reimbursed, partially reimbursed or deemed unrecoverable. This provides evidence of the total financial losses of consumers, as well as how existing reimbursement standards or requirements align with the ways in which consumers are being defrauded of their money.
3.3.8. Nature of authorisation (authorised vs unauthorised)
The dimension captures whether the transaction or payment was:
authorised by the user or payee, as defined by regulation, law, or contract between the customer and the financial service provider, or
not authorised, i.e. the procedure was bypassed, or the customer’s credentials were stolen or misappropriated.
This dimension therefore distinguishes financial scams and frauds based on whether the victim voluntarily authorised the transaction (due to deception or manipulation) versus cases where the transaction occurred without the victim’s knowledge or consent (typically due to account takeover or data misuse). This information therefore helps authorities to understand the circumstances that led to the victim being defrauded.
3.3.9. Status of entity
This dimension captures whether the entity was licensed to operate in the jurisdiction. For example, the entity could be one of the following:
fully licensed to operate for the activity related to the fraud, in the jurisdiction where the fraud took place;
licensed to operate for some activities in the jurisdiction, but not the activity related to the fraud;
not licensed to operate in the jurisdiction where the fraud took place, but is authorised in another jurisdiction;
or fictitious and thus neither licensed to operate in the jurisdiction where the fraud took place nor in another jurisdiction.
This can provide useful insight into the amount of fraudulent activity that is cross-border in nature and reveal how fraudsters may be exploiting ambiguity in existing laws and regulations.
3.3.10. The perpetrator
If such information is available, this dimension captures whether the fraudster was acting individually or whether they were part of a larger syndicate. This dimension may also capture if the perpetrator is operating across borders. For instance, organised criminal networks may be based in one jurisdiction, host servers in another and register domains in a third. If such information is known, this can provide additional evidence of the amount of fraudulent activity that is cross-border in nature and highlight the importance of intelligence sharing and co-operation by law enforcement across jurisdictions.
Table 3.1 summarises these ten dimensions with examples that can be used to describe and classify incidents of financial scams and frauds. These examples can be tailored to the national context.
Note: This is not an exhaustive list of examples within each dimension of financial scams and frauds targeting consumers.
The proposed typology aims to help public authorities that do not yet have a classification or typology in place monitor and analyse different types of financial scams and frauds. A classification template corresponding to the typology outlined above is included in Annex C. Public authorities are encouraged to use the classification template to ensure consistent classification along the ten dimensions. Depending on resources and the volume of fraud-related cases and/or consumer complaints, the classification template can be used for aggregated or sampled analysis, or as a conceptual reference to support data structuring at an aggregated level. Next, the data can be analysed to see where and how incidents ‘cluster’ along these ten dimensions. In other words, empirical evidence would be used to validate the typology and to identify the characteristics of the most common types of financial scams and frauds affecting consumers in a given jurisdiction.
Given existing research, this section provides three illustrative examples of likely fraud types that could emerge from empirical data.
This is how an incident involving phishing, wherein a fraudster sent an urgent email, could be classified using the proposed typology.
This is how an incident involving a fraudulent purchase could be classified using the proposed typology.
This is how an incident involving an investment scam could be classified using the proposed typology
Once public authorities have analysed the data to reveal the most common clusters, or types of financial scams and frauds targeting consumers within their jurisdiction, they can use the information to better understand how to tailor their financial consumer protection and financial education efforts. This information also provides an evidence base to measure fraud risks and provide support for the design and implementation of financial consumer protection and financial education policies, described in more detail in the next section.
In sum, classifications of financial scams and frauds can be helpful in assisting public authorities to systematically monitor and analyse different types of financial scams and frauds affecting financial consumers and retail investors. They can also assist in creating a typology that is consistently applied across public authorities, which then facilitates data sharing between different authorities. Additionally, typologies can direct consumer awareness and education efforts, as this information can be used by public authorities to focus their efforts on the most prevalent or severe types of financial scams and frauds. Lastly, such data can be leveraged to evaluate the impact of policies or interventions.
← 1. Definitions are available in Spanish at the following link: https://www.superfinanciera.gov.co/publicaciones/10109446/industrias-supervisadasinteres-del-vigiladoreportesindice-de-reportes-de-informacion-a-la-superintendencia-financieratablas-anexas-para-el-reporte-de-informacion-10109446/.
← 2. For more information about the European Banking Association’s Fraud Taxonomy, please see: What is the EBA Fraud Taxonomy? - EBA Association.
← 3. Jurisdictions may select the appropriate terminology depending on their national context.