offshore_leaks — ICIJ Offshore Leaks database

Description

The Offshore Leaks database of the International Consortium of Investigative Journalists (ICIJ): the offshore companies, foundations and trusts, the people and companies behind them, their intermediaries and their addresses, as revealed by the leaked documents of five investigations: the Offshore Leaks (2013), the Panama Papers (2016, from the law firm Mossack Fonseca), the Bahamas Leaks (2016), the Paradise Papers (2017, from the law firm Appleby and several corporate registries) and the Pandora Papers (2021, from several offshore service providers). The data cover offshore entities in over 200 countries and territories, over more than 80 years up to 2020. They are not complete records of the service providers, and the ICIJ notes that there are legitimate uses of offshore entities, and that the presence of a person or company in the database does not imply wrongdoing.

Each node is an entity (814,344: offshore companies, trusts and foundations), an officer (771,315: people or companies with a role in an entity, e.g. directors, shareholders and beneficiaries), an intermediary (26,768: e.g. lawyers, banks and agents that set up entities), an address (402,246), or another kind of node (2,989, e.g. companies from corporate registries). The 1,139 nodes that are listed both as officers and as intermediaries are single nodes. Each directed edge is a relationship as given in the database, e.g. from an officer to an entity of which it is a director (officer_of), from an intermediary to an entity (intermediary_of), from a node to its address (registered_address), or between nodes that the ICIJ found to be probably the same (e.g. same_name_as, similar, same_as); several relationships between the same nodes appear as multiple edges. The network contains all the 2,016,523 nodes and 3,339,267 relationships of the bulk download of the database generated on 9 September 2026.

The network has the following properties:

Nodes: node_id, the id of the node in the database; node_type, its kind (entity, officer, intermediary, address or other; officer; intermediary for the nodes listed as both); name; and, as given for the nodes (and empty otherwise): original_name, former_name, address, jurisdiction and jurisdiction_description (where the entity was incorporated), company_type, type, status, service_provider, internal_id, ibcRUC (a registration number), incorporation_date, inactivation_date, struck_off_date, dorm_date, closed_date (dates as given, mostly as DD-MON-YYYY), countries and country_codes (the countries the node is linked to, separated by semicolons), source (the leak and, for some, the service provider or registry), valid_until (how recent the data of the source are) and note.
Edges: rel_type, the type of relationship (e.g. officer_of); link, its description (e.g. "shareholder of", "director of"); status (e.g. "Resigned"); start_date and end_date, as given (in several formats), and start_time and end_time, the same as Unix time for the dates given as DD-MON-YYYY or YYYY-MM-DD (NaN otherwise); source, the leak (empty for 518,982 relationships).

Tags

Economic Social Offline Unweighted Multigraph Metadata

Citation

International Consortium of Investigative Journalists (ICIJ), "Offshore Leaks Database." (2026) offshoreleaks.icij.org

Upstream URL OK

https://offshoreleaks.icij.org/pages/database
https://web.archive.org/web/20260924142228/https://offshoreleaks-data.icij.org/offshoreleaks/csv/full-oldb.LATEST.zip

Upstream license

ODbL (database), CC BY-SA (contents), https://offshoreleaks.icij.org/pages/database

Networks

Tip: hover over a column header for a description.

Column legend

$\left<k\right>$: Mean degree
$\sigma_k$: Standard deviation of the degrees
$\lambda_h$: Largest eigenvalue of the Hashimoto (non-backtracking) matrix
$\tau$: Random walk mixing time
$r$: Degree assortativity
$c$: Global clustering coefficient
$\oslash$: (Pseudo-) diameter
$S$: Relative size of the largest component
NPs: Node properties
EPs: Edge properties

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
offshore_leaks	2,016,523	3,339,267	1.66	53.93	195.10	14330.09	-0.03	0.00	35	0.47	Directed	Unipartite	node_id node_type name original_name former_name address jurisdiction jurisdiction_description company_type type status service_provider internal_id ibcRUC incorporation_date inactivation_date struck_off_date dorm_date closed_date countries country_codes source valid_until note	rel_type link status start_date end_date start_time end_time source	44.1 MiB	66.3 MiB	69.1 MiB	79.0 MiB