Verification of beneficial ownership data
Verification after submission
Verification after submission should:
- ensure data is frequently checked;
- ensure data is kept up to date;
- ensure information suspected of being incorrect is investigated.
There are a number of general approaches to verification after submission, including checks after the publication of BO information. As with verification checks at the point of submission, multiple approaches can be deployed to complement each other and can mutually reinforce reliability and accuracy.
Ensuring data is frequently checked
Making beneficial ownership registers open and public
Making registers public allows for checking by the private sector, civil society, and the general public, both for accidental error and deliberate falsehoods. Research suggests that publishing data publicly can drive up data quality, as increased data use drives up the likelihood of inconsistencies or potential wrongdoing being identified. [6] In order for this to work effectively as a verification measure, mechanisms should be put in place to allow for reporting of errors, discrepancies and contradictory information. There are also a range of other benefits for the private sector, which are all expected to outweigh costs. [7]
Although there are no documented examples of harm as a result of public registers, [8] opponents of public registers frequently quote privacy issues as an argument against them. Governments should not disclose more data than necessary to provide meaningful oversight and transparency, and could include exemptions in the case of legitimate concerns.
Example: United Kingdom
In November 2016, Global Witness and a consortium of NGOs analysed 1.3 million companies in the UK’s Persons of Significant Control BO register. They were able to inform Companies House – the body overseeing the register – of over 4,000 companies with ineligible information. [9]
Sample testing/checking
Agencies responsible for BO registers can conduct in-depth investigations of samples of the data or require external parties to do so. These tests provide a deterrent to companies against submitting wrong information. Sample testing may not be a very effective verification mechanism and can be quite resource intensive. This can be mitigated by using a risk-based approach to sample testing.
Example: Denmark
To ensure that BO information in the Central Business Register (CVR) is accurate and current, the Danish Business Authority (DBA) started manually checking 500 companies and their registration of beneficial owners in 2019. [10]
Ensuring data is kept up to date
Require updates to the information in case of changes
BO changes should be required to be updated swiftly following changes. Specifying a short and defined time within which to submit any changes to a register ensures BO information stays up-to-date. Public registers can also publicly display when information is out of date to alert data users. Being required to submit frequent updates to the register has the potential to raise compliance costs, which should be factored into verification system design.
Require confirmation of existing information
Disclosing entities should check and confirm on a regular basis (at least annually) that their BO info is accurate and up-to-date. This can be integrated with existing business processes (e.g. submitting annual returns). Without other verification checks, however, this measure is ineffective.
Example: Ukraine
In order to ensure constant updating of information on beneficial owners, the Ministry of Justice of Ukraine issued Order No. 2824/5 “On Making Amendments to Certain Forms of Applications in the Field of State Registration of Legal Entities, Individual Entrepreneurs and Public Organization” in 2018, which obliged the companies to update information on their ultimate beneficial owners when changing any information with the Unified State Register, or to confirm the information held is still correct. [11]
Ensuring information suspected of being incorrect is investigated
Require reporting of suspicious entries and activities
Bodies dealing with BO data should be required to report suspicious submissions and activities to the appropriate bodies, and they should be mandated to investigate these (e.g. private sector conducting due diligence could report to the Financial Intelligence Unit (FIU) for reports related to money laundering). It is important that the FIU’s are appropriately resourced to be able to investigate reports (see example).
Example: United Kingdom
From January 2020, sectors that fall under anti-money laundering and counter terrorism-financing AML/ CTF regulations are required to report discrepancies between beneficial ownership information available at Companies House, and information that they obtain through their own compliance checks. [12]
Example: The Netherlands
An estimated €16 billion is laundered through the Netherlands each year. While obligated entities reported 60,000 suspicious transactions in 2018, the FIU only deemed 15,000 of those as actually suspicious, but is suspected of only being able to investigate far fewer, due to a lack of (human) resources. [13]
Red-flagging
Systems can be set up to detect patterns associated with legal vehicles being used for illicit purposes. This is likely to be highly context-specific. These systems will be easier to set up in digitised systems with BO information as structured data, and could adopt AI and machine learning technologies. There is a risk that when additional red-flagging checks are added and BO information is cross-checked with additional registers, the number of entries falsely flagged as suspicious will also grow, decreasing its utility. It is therefore important to also consider mechanisms to reduce these errors, and to introduce a lightweight and rules-based business process that responds to these discrepancies.
Example: Ukraine
In Ukraine, the working group on verification “Up to 100%” has proposed a number of verification systems that raise automatic red flags based on known structures used for illicit purposes. For instance, in Ukraine it is common to list a factory worker as a BO. The proposed system would automatically raise a red flag for investigators when someone is listed as a BO of a profitable company while tax data shows that person earning a wage significantly lower than what could be expected from a profitable company owner. [14]
Most beneficial ownership disclosure regimes will deploy a number of these verification mechanisms, which is by no means an exhaustive list, but all fall broadly within these three approaches. No single approach is better and ultimately their success will be highly dependent on the context in which it is deployed and what other checks are in place. Countries should therefore take a holistic and comprehensive approach to verification, taking a risk-based approach and bearing in mind the overarching aims of the verification system as means to an end to facilitate data use and, in turn, policy impact.
Footnotes
[6] OpenOwnership, “Briefing: The case for beneficial ownership as open data”. July 2017. Available at: https://www.openownership.org/uploads/briefing-on-beneficial-ownership-as-open-data.pdf [Accessed 20 April 2020].
[7] Ibid.
[8] OpenOwnership, The B Team and The Engine Room, “Data Protection and Privacy in Beneficial Ownership Disclosure”. May 2019. Available at: https://www.openownership.org/uploads/oo-data-protection-and-privacy-188205.pdf [Accessed 20 April 2020].
[9] Global Witness, “The Companies We Keep”. 2016. Available at: https://www.globalwitness.org/documents/19400/Briefing_The_Companies_We_Keep.pdf [Accessed 20 April 2020].
[10] FATF, “Best Practices on Beneficial Ownership for Legal Persons”. October 2019. Available at: https://www.fatf-gafi.org/media/fatf/documents/Best-Practices-Beneficial-Ownership-Legal-Persons.pdf [Accessed 20 April 2020].
[11] Based on “Concept of a mechanism for verifying the reliability of information on UBO” shared with OpenOwnership by the “Up to 100%” verification working group, as well as discussions with the working group members in February 2020.
[12] HM Treasury, “The Money Laundering and Terrorist Financing (Amendment) Regulations”. 2019. Available at: http://www.legislation.gov.uk/uksi/2019/1511/made/data.pdf [Accessed 20 April 2020].
[13] Trouw, “Belastingadviseurs: ‘Overheid is te slap tegen witwassen’”. 9 February 2020. Available at: https://www.trouw.nl/economie/belastingadviseurs-overheid-is-te-slap-tegen-witwassen~b0f40eff/ [Accessed 20 April 2020].
[14] Based on “Concept of a mechanism for verifying the reliability of information on UBO” shared with OpenOwnership by the “Up to 100%” verification working group, as well as discussions with the working group members, in February 2020.