Global Coalition to Fight Financial Crime position paper: Advancing interoperable and usable beneficial ownership information

Position paper

The Global Coalition to Fight Financial Crime (GCFFC), as part of its objectives to promote more effective information sharing between public and private entities, and to propose mechanisms to identify emerging threats and best practice approaches to more robust controls against money laundering, believes that all actors fighting financial crime should have instant access to high quality, highly-usable beneficial ownership (BO) data.

The value of BO information for fighting financial crime has been widely recognised by international standard-setting bodies, including the FATF, OECD, IMF, the World Bank, and the United Nations. This GCFFC paper outlines key considerations to maximise the potential of this information as an anti-financial crime tool. Taking a deep dive into a technical aspect of beneficial ownership transparency (BOT) reform implementation, it outlines key considerations for building technological systems that collect and share structured, interoperable and usable BO data. Building on a growing body of knowledge on effective reform, including the analysis of existing BO data structures and the development of the Beneficial Ownership Data Standard (BODS), the paper is primarily relevant for policy makers and government agencies responsible for the implementation of BOT reforms. It provides a set of key considerations to support jurisdictions in standardising BO data collection and publication to ensure it is well-structured and interoperable. To do so, technological systems should involve, at minimum:

1. Collecting and publishing structured BO information on three elements: 1) individuals, 2) legal entities or arrangements, and 3) the relationship between the individuals and entities or arrangements, including the nature of the interest(s);
2. Collecting and sharing information on intermediary entities and arrangements involved in the relationship;
3. Enabling the use of reliable identifiers and widely used data formats across multiple databases;
4. Using standardised data formats for common types of fields with comprehensive documentation for users to make data more reliable and usable;
5. Collecting and publishing information about change over time; and,
6. Providing a variety of ways to access and process data, including via an Application Programming Interface (API) and in bulk, which will enable a wider range of possible analyses.

Solid international cooperation mechanisms and some level of alignment between domestic policy and legal frameworks is also required to enable the interoperability of data, facilitate the interconnection of information from domestic registers and understanding of transnational ownership networks.

Context

Evidence from an increasing number of jurisdictions implementing BOT has informed a better understanding of what constitutes effective reforms and what challenges data users face – from financial institutions, financial intelligence units, tax authorities and law enforcement to investigative journalists, watchlists and other service providers that contribute to enhancing the work of these actors – to use BO information to fight financial crime effectively. Both domestic governments and international institutions are developing policy and technological solutions to these challenges. For example, the European Union’s (EU) 2024 AML package seeks to ensure that all relevant stakeholders can readily access and effectively use BO information made available through member states’ BO registers. The package highlights the need to ensure data from different jurisdictions is collected in a standardised format. Standardising the collection of information and structuring this information according to a common standard will facilitate the interconnection and combining of information within and between countries – or data interoperability – and to make data easier to use, especially to address the transnational nature of financial crime. The Commission will prescribe this format within one year of the date of entry into force of the package.

The GCFFC comprises different actors engaged in fighting financial crime: law enforcement agencies, obliged entities, and civil society organisations. In 2021, the GCFFC built on the collective expertise of its members to outline various legislative, regulatory and technological features needed to produce high-quality BO data that could be used by the widest possible range of anti-financial crime actors. While various countries have made significant progress in implementation since 2021, not being able to readily connect BO information to other datasets remains a key barrier to usability for data users. Building on the 2021 paper, this document explains the need for and key technical considerations for ensuring BO information is available as structured, standardised and interoperable data.

What makes beneficial ownership data usable to fight financial crime

BO data is part of an ecosystem of information that users need to connect to in order to trace the proceeds of crime and build a complete picture of the networks of individuals, legal entities and arrangements (collectively: corporate vehicles) involved. This ecosystem can include a large variety of types of information, such as other information related to corporate vehicles and individuals, financial and transaction information, information on physical assets such as real estate, public procurement information, adverse media, the track record in a given sector of activity, or information about political exposure.

All users need to bring information from different domestic and international sources together for a range of use-cases to fight financial crime. Many users currently face barriers to doing so, including effectively accessing the type of information they need, and being able to process it in ways that allow those fighting financial crime to find the answers they need to often complex questions. To enable this, it is essential to consider users’ needs and experience, including challenges faced in accessing and processing BO data.

BO data users are currently facing a range of challenges in effectively using data to fulfil their anti-financial crime goals. A number of studies – for example United Nations surveys of around 90 countries, a review of the 4th EU AML directive, World Bank country case studies in BOT implementing countries, and primary research conducted by GCFFC member Open Ownership – have started documenting the most common needs and challenges. They are summarised below.

• Connecting BO data: making sense of data from multiple sources often involves entity resolution. That is, identifying whether two or more data elements from several sources correspond to the same real-world element. For example, to establish whether a set of records from a BO register and a sanctions list, all refer to the same individual, despite certain discrepancies. The lack of effective mechanisms to enable entity resolution significantly increases the time and resources that a wide range of users spend to connect different datasets, sometimes involving manually matching individuals or companies from different sources. This is often due to the lack of reliable unique identifiers associated with each individual or corporate vehicle across multiple BO and non-BO datasets. It is also due to the lack of consistency in the range and format of accessible information or fields across various BO registers (e.g. full name, year, month of birth, nationalities, phone number etc.) that users can refer to in the absence of unique identifiers. Finally, the absence of information – for example, information on intermediary corporate vehicles, specific corporate vehicle types or BO information from specific jurisdictions – means that users can face dead-ends in investigations. All these challenges reduce the interoperability of BO information.
• Processing large quantities of data: to support obliged entities in spotting risks more efficiently to meet their regulatory obligations, third-party providers of AML/CFT products and software solutions and other data service providers propose tailored products with enhanced analysis and functionalities that are not necessarily available on BO registers. These services currently play a central role in addressing the lack of interconnectivity of data sources. To be able to provide functional services allowing users to access and analyse large quantities of information, often in real time, these service providers need to access information through an API. Other users, such as academics seeking to identify trends and risk patterns using the whole dataset will need to download data in bulk to carry out their analyses.
• Querying and analysing information in more advanced ways: those who use BO data to fight financial crime often need to be able to conduct more advanced queries and analyses than the basic search functionality offered by many registers. This may range from one specific person or type of corporate vehicle, to multiple individuals or entities with specific characteristics (e.g. a politically exposed person, people of a certain nationality, entities registered in a certain jurisdiction). When data is not collected and published in a highly organised and structured way, and registers only offer limited search functionality, a range of users cannot query the data effectively. Structuring data creates information that is predictable. This provides a basis to enable other features which support users to analyse information, such as capturing and representing change over time. Actors fighting financial crime need historical data to be able to monitor and identify specific changes over time which may indicate risk.
• Relying on registers: well-structured data improves data reliability by improving its accuracy. Systems that collect information as structured data, as opposed to as free text or PDF files, can reduce accidental errors and enable compliance of those disclosing BO data. Collecting and publishing structured data across multiple datasets also makes it easier to then make data interoperable.

Developing systems tailored to user needs

Solid technological systems require comprehensive policy and legal frameworks. These frameworks are an essential foundation for usable and interoperable data. While each context is different, some level of alignment across domestic frameworks is required to enable the interoperability of data and interconnectedness of registers. For example, UN surveys underline the importance of unified robust definitions of BO in legislation. Ensuring that effective enforcement mechanisms exist across countries to enable the availability of data, and ensuring that information from different sources is verified to a similar level of accuracy is equally crucial to create the conditions for BO data to be effectively used and combined. Building on a growing body of knowledge on effective reform, including the analysis of existing BO data structures and the development of the Beneficial Ownership Data Standard (BODS), the following considerations are informed by the latest evidence to maximise BO data’s usability. To collect, store, and share BO information, technological systems should, at a minimum:

1. Collect and share structured BO information about a minimum of three elements: 1) information on the individuals disclosed as part of a BO declaration, including the beneficial owners and any other relevant parties (e.g. nominees), 2) information on the legal entity or arrangement these individuals are disclosed as the beneficial owners of, and 3) information on the relationship between the individuals and the entity or arrangement disclosed in the BO declaration, including the nature of the interest(s) (e.g. voting rights, percentage of shares, or nominee arrangement).
2. Collect information on intermediary entities and arrangements: Multiple legal entities or arrangements can be involved in the relationship between a given corporate vehicle and its beneficial owners. Collecting and sharing information about these intermediaries assists users in building better pictures of BO networks and can help users uncover complex corporate structures. This enables more comprehensive tracing and identification of relevant individuals and assets involved in financial crime. Requiring the disclosure of at least one intermediary corporate vehicle – both legal entities and arrangements – in cases of indirect beneficial ownership makes the visualisation of complex networks of beneficial ownership possible. Where possible, shareholder information can be used to complement BO declarations, aid compliance and support the verification of BO declarations.
3. Enable the use of reliable identifiers and widely-used data formats across multiple databases: Examples of identifiers include registration numbers issued to legal entities by corporate registers, national identity numbers, or an internal identifier generated for a beneficial owner by a BO register. Using identifiers issued by authoritative registration bodies – such as the Global Legal Entity Identifier Foundation (GLEIF) – provides reliability for users to check that an individual, entity or item exists in the real world. These can be used across multiple information sources (e.g. BO registers, asset registers, obliged entities’ information systems etc.). The more these are used across different information sources, the easier it is to connect elements across them. Identifiers are essential to make data interoperable. Choosing a well-used format for exchange of data (e.g. JSON) is also important to maximise the potential for data to be easily linked.
4. Use standardised data formats for common types of fields, with comprehensive supporting documentation: Using standardised formats for common data fields is a key part of structuring data. For example, a date could be recorded in various ways – 2022-07-11 or 11 July 2022. By explicitly defining a specific data format – for example the International Organization for Standardization (ISO) 8601 standard, which reads (YYYY-MM-DD), it becomes clear that 2022-07-11 means the 11 July 2022 and not 7 November 2022. Publishing and maintaining guidance for users of BO data enables them to correctly discern its meaning. For example, if "beneficialOwnerDate" in a dataset is given as 2022-07-11, users of the data need to understand more than just the calendar placement of the date; they need to know whether it is the date that an individual became a beneficial owner, that they were declared as a beneficial owner, or something else entirely. Where information on both legal and beneficial ownership is collected and published in a single register, aligning the types and format of fields used in both types of information will help build full pictures of ownership networks and better address financial crime.
5. Collect and share information about change over time: Collecting structured data, using reliable identifiers and standardised formats is a prerequisite for building a database that provides a comprehensive record of changes in BO information over time. It is essential to capture information about who (through reliable identifiers) knew and declared what (standardised fields, clear relationship between people, entities and arrangements linked to identifiers, documentation on the rationale and nature of the change) and when (standardised date formats).
6. Provide a variety of ways to access and process data to enable a wider range of possible uses in accordance with policy and legal frameworks: In some cases, users need to process large quantities of data to identify financial crime risks and therefore require access to data in bulk or via an API. Some targeted work on financial crime will require searching through a BO register according to specific criteria (e.g. jurisdiction of incorporation, beneficial owner nationality, number of times an individual is declared as a beneficial owner etc.). Others will need to visualise complex ownership networks in a graph, or network, format. Having a nuanced understanding of the different ways BO data is used from the outset of reforms is essential to design laws and systems that enable the greatest variety of use cases.