The Tax Justice Network’s Financial Secrecy Index (FSI) latest edition was published on January 30th, 2018. As always, the FSI offers a ranking of the world’s largest contributors to global financial secrecy as well as narrative reports for most of the most famous secrecy jurisdictions, describing their “tax haven” history and summarizing the FSI findings.
Furthermore, the FSI offers detailed database reports for each of the 112 covered jurisdictions, describing their legal framework on financial transparency and available loopholes. These detailed reports (which contain the corresponding description, source and date for each answer) provide the explanation for the secrecy score of each country as well as additional details on a jurisdiction’s transparency. The secrecy score is one of the two components driving the FSI ranking of each jurisdiction.
The latest edition of the FSI offers the detailed reports not only online and for free (as always), but now also in open data format (downloadable as Excel documents). The FSI also offers two interactive maps to play with the data.
The availability of FSI data in open data format underlines TJN’s commitment to full transparency. We are already beginning to reap the benefits. Committed experts have started to look in-depth at our data and advised us of glitches and other issues that require fixing.
The glitches we have acted upon and ‘fixed’ do not affect the secrecy score of a jurisdiction and the FSI ranking remains the same. However, we cannot rule out that more changes will be required in the future. The FSI team spends considerable amount of time assessing each question (especially those affecting the secrecy score of a country). Contested issues are discussed in the team, and final audits and cross-checks of information take place before publication. Still, the detailed database reports include up to 186 questions and answers for each of the 112 jurisdictions. That’s a lot of questions and, of course, we are human and not immune from error! One strategy to minimise the risk for errors is to release all of our data and sources to the public, for others to use it, to check it and also to question it, and thereby improve it if necessary.
The FSI database is a living document and we have now published a new layout version with some additional fixes (e.g. rephrasing some questions for greater clarity, removing some superfluous items from display, etc.). Once again, we would emphasise that the changes do not affect the secrecy score or the ranking of a jurisdiction.
We welcome others to look in detail at the FSI data and report any question or comment they may have. Those interested in doing this should consider:
-Focus on questions with an asterisk *, because this indicates that the question feeds into the country’s secrecy score and thus affects the jurisdiction ranking.
-The FSI applies the lowest common transparency denominator. This means for example that if a country has many types of companies that are very transparent (e.g. they all must register their beneficial owners), but one type of company is very secretive, then the secrecy score of the country will be based on the most secretive type, because we understand that anyone engaged in illicit activities would likely use the secretive type. Some may argue that this is misleading, because the secretive type may be the exception, but not the rule. However, this is precisely what the FSI looks for: exceptions, loopholes, ring-fences, gaps or anything that could be exploited to abuse secrecy or tax rules, despite general rules. These exceptions prove that countries – and not the FSI – can be misleading: the apparent high statutory tax rates, or comprehensive transparency measures are usually undermined by fine print loopholes.
All comments received will be analysed and responses given to the commentator, but changes would only apply if a source is provided and the change is consistent with the FSI criteria (e.g. notifying us about the “general rule” or a “best case” will not suffice to change an answer if the loophole or exception still applies). While we cannot guarantee to constantly change the published data, all relevant changes that have not been (publicly) addressed will be implemented in 2019/2020 when we undertake the biannual reassessment of the FSI.
You can access the updated FSI database reports here, and the updated Excel extracts, here. Anybody interested to compare against the previous versions, a zipped archive of the database reports published on 30 January 2018 can be accessed here (to open the xml files, please extract the entire zip in a folder and open with internet browswers. For technical reasons this might only work with PCs, not with MACs – sorry!), and of the old Excel extracts, here.