We often talk about sanctions in the context of efficiency or individual cases. Who’s who in the sanctions space has, as a result, become elusive when there are characteristics, data sets, and methods of revealing sanctioned entities and their networks.
Let’s explore three common steps in the research process behind investigations in the space of anti-money laundering solutions, from the design of research methods to the collection and analysis of information.
Research Design
Ontology
The first step in researching who’s who in the sanctions space is knowing how to describe entities and relations between them. In industry, we describe our domain and its objects in an ontology.
An ontology is a formal naming and definition of categories, properties, and relations between concepts, data, or entities.
I recommend using the Follow the Money ontology.
Follow the Money
Follow the Money is an ontology that is commonly used in anti-corruption investigations. Their ontology defines entities relevant in such investigations, like people or companies, and helps investigators create a graph of relationships.
One of the most common tasks for an investigative reporter is to work out if one thing is the same as another.
Is the person who owns a sanctioned company the same guy who is a mayor in my city?
Many facts might provide hints that allow us to establish if he is “our guy.” Do the individuals have the same name, date of birth, nationality, passport number? Do we know about links to other individuals or companies in both data sets?
When you have a clearly defined ontology, you can turn representations of objects and events into complex data and compile and cross-reference these hints.
The Follow the Money ontology consists of schemata—i.e., object types—that exist as an inheritance hierarchy rooted in things and events.
- Things describe real-world objects like people, companies, assets, court cases. You can think of things as entities.
- Intervals are business interests, court cases, sanctions, and transactions, and are useful for linking entities together over time.
Build a Taxonomy
Once you have an ontology that structures your domain and its objects and relations, you can build a taxonomy to describe and group your information.
While an ontology is a formal representation of knowledge that defines concepts and relationships within a domain in a structured way, a taxonomy concerns the classification of entities based on their characteristics.
Taxonomies are used to group types of entities into categories. For example, when we talk about associations and relationships in the sanctions space, we need ways of describing the type of entity or relationship.
An associated entity could be an entity type: person, organization, or vessel. If we know that our entity is a person, we can classify their relationship to the sanctioned entity.
Taxonomy of Relationships
| Relationship | Definition | Example |
|---|---|---|
| Employee | The associated person is employed by the sanctioned individual | A nuclear engineer working for a sanctioned company |
| Owned or controlled by | The associated organization is owned or controlled by the entity it connects with | OOO NOVATEK-PERM is a subsidiary owned by sanctioned OAO NOVATEK |
| Family member | An associate is a spouse, child, parent, or other relative of a sanctioned person | Asma al-Assad is the spouse of Bashar al-Assad |
Having a taxonomy is crucial to investigations. A good taxonomy lets investigators describe and filter the information they’re collecting.
Collect Data
Global sanctions regimes vary, as some countries implement and enforce the sanctions of international organizations such as the EU and the UN, while others develop their own sanctions programs and lists.
The first step in researching sanctioned entities is knowing which ones appear on an official sanctions list.
Key Sanctions Lists
Government and financial authorities maintain targeted sanctions lists. These lists are often available online and evolve quickly.
- UN Sanctions List (UNSC Consolidated List): Consists of individual targets and entities or groups targeted by UN Security Council resolutions.
- US Sanctions: The Office of Foreign Assets Control (OFAC) maintains:
- Specially Designated Nationals (SDN) List (individuals and companies)
- Consolidated Sanctions List (additional sanctions)
- UK Sanctions: Enforced by the Office of Financial Sanctions Implementation (OFSI).
- Australia Sanctions: Implemented by the Australian Sanctions Office (ASO) and DFAT.
- Canada Sanctions: Under the Special Economic Measures Act (SEMA) and the Justice for Victims of Corrupt Foreign Officials Act (JVCFO).
- Japan Sanctions: Issued under the Foreign Exchange and Foreign Trade Act (FEFTA).
- China Sanctions: Includes the Unreliable Entity List (UEL), the Anti-Foreign Sanctions Law, and the Blocking Rules.
Data in Key Sanctions Lists
Sanctions lists are useful primary sources to identify who’s who in the sanctions space and as a stepping stone to identifying associates and enablers.
Common information about sanctioned entities includes:
- Name
- Alias
- Nationality
- Passport number
- Country of birth
- Reason for being sanctioned
Sanctions lists are usually structured (XML, CSV, HTML), but may also be unstructured PDFs requiring manual extraction.
Example extract (UNSC Consolidated List):
Name: ABDUL LATIF MANSUR
DOB: Approximately 1968
Nationality: Afghanistan
Title: Maulavi
Designation: Minister of Agriculture under the Taliban regime
Aliases: Abdul Latif Mansoor, Wali Mohammad
Listed on: 31 Jan. 2001 (amended multiple times)
Other information: Taliban Shadow Governor for Logar Province...
These data points can help confirm if the person on a sanctions list is the same as a person in a company register.
Limit of Primary Sources
Investigating whether an entity is sanctioned is relatively straightforward. The challenge is identifying associates or enabling entities.
For example, a company may not appear on a sanctions list, but if a sanctioned entity owns 51% of it, both are sanctioned (the OFAC 50% Rule).
Sanctions lists are excellent primary sources, but investigations require a holistic approach.
Data Enrichment
Other sources are used to enrich the data from sanctions lists, such as:
- Ownership registries
- Land registries
- Leaked documents
- Mentions in media
- Aggregated databases
Case study:
Roman Abramovich was sanctioned after Russia’s invasion of Ukraine. Using the Oligarch Files (leaked data from MeritServus), investigators revealed he had covertly funded the Dutch football club Vitesse Arnhem via offshore entities.
The Guardian’s investigation identified six companies involved in channeling €117 million to the club.
Entity Resolution
In AML and illicit finance, we strive for a source of truth. Enriching data requires entity resolution—identifying records across sources that refer to the same entity.
Entity resolution can be manual (comparing names, dates of birth, etc.) or automated using techniques like:
- Fuzzy matching
- Merge-purge
- Data matching
- Identity resolution
Through entity resolution, we can link records correctly and enrich our understanding.
Network Analysis
Data dumps are rarely useful on their own. Network analysis helps investigators extract insights about:
- Network risks
- Broader trends in illicit finance
For example, visualizing the network of Abramovich’s companies and connections clarifies how offshore structures circumvent sanctions.
Recap
Every investigation requires four simple steps to be successful:
- Decide how to describe objects and relationships (ontology).
- Set up a system to classify objects (taxonomy).
- Identify and collect information from reliable sources.
- Enrich data, perform entity resolution, and analyze networks to find actionable insights.