Areas for improvement
The UK’s public register was pioneering in many ways, and while many parts of the register set new standards, there are areas where there is room for further improvement.
The UK only requires beneficial owners to report themselves as such if they control 25% of the shares or voting rights in a company. This threshold creates a risk that significant interests in a company may not be represented in the register; after all, 20% of a company can be a large and lucrative stake. In extractives, for instance, where company ownership can be extremely profitable and corruption risks are very high, even a 1% stake in a company is a relationship worth reporting. Global Witness has identified multiple examples where owning as little as 10% of a company has raised serious red flags.
The Nigerian Ministry of Justice identified this threshold as one of the key challenges in the UK register, stating that “there is a strong argument for reduction of the threshold as it is suspected that this is being exploited by some businesses to avoid full compliance with the reporting rules.” In Ghana, the amended Companies Act does not set any threshold percentage for defining beneficial ownership.
Countries should consider defining a beneficial ownership threshold lower than 25%.
Another lesson learned from the UK is the potential challenges raised by the banding of ownership stakes. Beneficial owners report their stake as existing on a range from 25% to 50%, 51% to 75%, or 76% to 100%. One issue is that this is imprecise, and no additional burden is presented by asking the filer for a specific percentage. Another is that it makes it challenging to compare data across jurisdictions, unless that data is modelled in precisely the same way, with matching bands.
Beneficial owners should be required to report their holding of shares or voting rights in exact percentages.
One limitation of the register is that by only providing the name and month and year of birth it can make it difficult to ascertain which records refer to the same person when there are many people with the same name. This is a challenge both to compare within the UK registry where individuals are beneficial owners of more than one company and to be able to compare against other data sets.
One solution would be for the register to allocate unique identifiers to individuals that would allow the registry to link records where individuals are beneficial owners or company officers (e.g. directors or secretaries) of more than one company and help users of the data to match records against other data sets. This unique identifying number should be specific to the database, not a piece of personal data such as a personal ID or passport number. Companies House already links some records for company officers across companies, a functionality which should be expanded to include all company officers and beneficial owners.
The use of unique identifiers to connect individuals within corporate research has been shown to have been enormously powerful. In 2015 Global Witness published a ground-breaking report revealing the network of military elites, US-sanctioned drug lords and crony companies controlling Myanmar’s multi-billion dollar jade industry. Without unique identifiers to connect individuals, particularly in a context where some names are very common, this analysis would not have been possible.
Registers should use unique identifiers in addition to personal data such as name and month and year of birth.
Data validation at data entry
How users are allowed to enter information into the registry can have a huge impact on the quality and accuracy of the resulting data. For example, people submitting information to the UK register were asked to type their nationality into the relevant field, resulting in over 500 spellings of ‘British’ and 10 beneficial owners listing their nationality as Cornish (a county in South West England). The resulting data is poorer quality, more difficult to disambiguate, and generates lower confidence from users. This particular issue is easily resolved by replacing the free text input with a choice from a pre-populated list of recognised countries, which Companies House now intends to implement imminently to a new Companies House and HMRC service for incorporating certain company types. This will then be rolled out to other incorporation services later this year before being extended to other filings in 2018. This same approach could also be used for other standardised, canonical data fields, such as UK postal address validation.
The lack of data validation on the date of birth field also allowed 2,160 beneficial owners to provide their date of birth as 2016 and others as far into the future as 9988. Following the findings of the data analysis exercise in November 2016, Companies House has now included a prompt within the PSC Register when users provide a date of birth which is below age 16 or over 100 and preventing people entering an age over 110 through the online system.
Basic data validation systems such as multiple choice fields should be used to improve data quality.
One of the most significant weaknesses of the UK register is that the data submitted is not verified, it is solely self-reported data from companies. This means that data can be submitted that does not comply with the requirements of the register or is inaccurate.
Analysis of the UK register found multiple examples of potential non-compliance, including listing companies based in tax havens as beneficial owners or reporting looped ownership where companies appear to own themselves. In February 2017, Companies House identified 4,500 companies that had reported a company located in a tax haven as their beneficial owner. Companies House decided to run a pilot targeting approximately 250 suspected non-compliant companies, alerting them to their non-compliance and asking them to review their details after which approximately 70% of these companies corrected their PSC entries in the register.
Companies House should expand these initial pilots to undertake regular systematic analysis of the registry to identify and pursue potentially non-compliant companies.
Systems should be put in place to identify potential non-compliance and pro-actively pursue companies that report non-compliant data.
To verify or improve the accuracy of the data there are a range of approaches that could be taken, including:
Requiring entities that conduct customer due diligence to file reports to the register, regulators or law enforcement if the beneficial ownership data they find does not match the public register. This has been proposed by the European Parliament for the revision of the Anti-Money Laundering Directive.
Requiring the submission of proof of identity or documented proof of ownership / control of the company. Denmark - which has recently also set up a public register of beneficial ownership for companies - requires beneficial owners to submit a scanned copy of their passport or other national ID, limiting the possibilities for false registrations.
Cross-checking data against other government data sets.
Establishing systems for members of the public to easily highlight or report suspected inaccurate data in the registry.
It is important to note that all but one of these measures are only possible when the information is available as structured data, and that two are only possible when that data is available to members of the public to use and re-use.
The resources for these activities should be proportionate to the scale of the risk of abuse and the impacts caused by abuse of corporate vehicles through hidden beneficial ownership. The UK’s AML NRA identifies that 70% of money laundering cases being investigated by HMRC used company structures for money laundering, shifting a total of £800 million. Much of the costs for identifying non-compliance with the register could be recovered from the proceeds of financial penalties.
As part of the implementation of the Fourth Anti-Money Laundering Directive, the UK is considering what additional measures may be needed to improve the quality of the data.
A range of systems should be established to improve data quality or verify data.
 See Annex for the rules on how a PSC is defined.
 Global Witness, ‘Assessment of EITI Beneficial Ownership pilots’, p.7, March 2015.
 Federal Ministry of Justice of Nigeria and IBLF Global, ‘Improving the business environment in Nigeria through transparency in the management of beneficial ownership - a policy brief’, Feb 2017.
 Republic of Ghana, ‘Ghana EITI Beneficial Ownership Road Map’ 1/10/2016.
 Open Corporates, ‘How open company data was used to uncover the powerful elite benefiting from Myanmar’s multi-billion dollar jade industry’, 27/10/2015.
 November 2016 analysis of PSC register data by Global Witness, DataKind UK, OpenCorporates, Spend Network and OCCRP
 November 2016 analysis of PSC register data by Global Witness, DataKind UK, OpenCorporates, Spend Network and OCCRP.
 Meeting with Companies House, Cardiff, September 2017.
 European Parliament, ‘Report on the proposal for a directive of the European Parliament and of the Council amending Directive (EU) 2015/849 on the prevention of the use of the financial system for the purposes of money laundering or terrorist financing and amending Directive 2009/101/E’, 9/3/2017.
 Lexology - Bech-Bruun, “Mandatory Registration of Beneficial Owners”, 31 May 2017.
 HM Treasury/ Home Office, ‘UK national risk assessment of money laundering and terrorist financing’ October 2015.
 HM Treasury, ‘Consultation outcome - Money Laundering Regulations 2017’, 26/6/2017.