This is a short series of blogs on transfer pricing documentation: what I have seen over the years, how it does (not) align with OECD documentation requirements, and thoughts on possible improvements. I focus on practicalities.
Why do we need TP documentation?
One answer is because many countries’ laws require it. More interesting is the reasons given by the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 2017 (the “OECD TPG”). According to §5.5, the objectives of TP documentation are:
- to ensure that taxpayers give appropriate consideration to transfer pricing requirements in establishing prices and other conditions for intercompany transactions;
- to provide tax administrations with the information necessary to conduct an informed transfer pricing risk assessment; and
- to provide tax administrations with useful information in an appropriately thorough audit of the transfer pricing practices of entities, although it may require additional information as the audit progresses.
The first objective fascinates me. I do not think of it as an evidence exercise, as much as I think of it as fundamental understanding of human nature. When we are asked to commit to things in writing, we focus, we consider, we debate. In the process, we go through things far more thoroughly than we ever would have, had we not been required to put into writing our reasoned opinions about what it is that we have done.I think it has been wise of the OECD to include this, often underappreciated, goal of transfer pricing documentation.
However, the first objective also carries within itself its very own downfall. The more template driven, resource restricted and budget focused TP documentation becomes, the less likely it is that the documentation will achieve this primary objective, and the more likely it becomes that it will fail in its other objectives as well. I will add to this one more thing: the further away the preparer of the TP documentation sits from the decision makers behind the intercompany transactions, the lesser the “appropriate consideration” will be (hence my apprehension for outsourcing TP documentation).
Finally, and seen from a taxpayer’s perspective probably nirvana, the ultimate objective of TP documentation is to make a tax inspector put it down after reading it, thinking “I am done here, no adjustments are necessary”. Needless to say, the chances of that happening are not that great when a transfer pricing file a) looks like an extensive copy-paste exercise and b) hence is poor providing in taxpayer specific facts and circumstances.
How to make TP documentation
One of the best outcomes from the OECD BEPS project has been a strong global standardisation of TP documentation. Basically, the OECD took the EU Masterfile concept from the early 2000’s and added a whole lot about intangibles to it. BEPS Action 13 lead to a new chapter V on Documentation and several annexes to the OECD TPG. Annex I and Annex II describes exactly what information should be included in the master file and in the local file.
This begs the question, why does one see so many TP documents that do not follow these instructions step by step. There are different possibilities; I will mention two:
- there were already TP documents in place before BEPS (and before the EU Master file). Rather than wholeheartedly adopting the new guidance, the documentation preparers preferred to continue working with the old files, adding stuff they deemed necessary as they went along (and rarely subtracting material that has become superfluous).
- I suspect a difference between accountants and lawyers in applying instructions. Lawyers tend to stick very closely to the letter of the law when it comes to compliance, accountants seem to take a more interpretive approach. Through experience, I believe one may also observe that where lawyers tend to weigh every word, accountants may think more in concepts. I have posted about the use of words before. As the vast majority of TP outsourced documentation tends to be prepared by big 4 firms, I suspect one may also expect more of an accounting than a legalistic approach.
Does it matter that TP documents do not follow Annex I and II step for step? I think it does.
- The greatest danger is the increased chance of omissions. If I asked you to do steps 1 – 10, and you start with 1, 2, 4a, 8, 6a, 10a, 10c, 10b, 3a, 7, 6b, 4b, 3b and 6c, chances are significant that you do not notice that you have omitted 5 and 9. Likewise, if my step 5 is spread over your steps 4a, 10c and 3b, chances are that you covered my step 5 with things I did not ask for whilst still lacking things I did ask for. Why this unnecessary complexity?
- A seemingly lesser danger is the provision of superfluous information. I have seen TP documentation with vast amounts of irrelevant corporate history (i.e. having no influence on the transactions under discussion), enormous industry analyses (overkill) and pages and pages of copies (or worse, outdated summaries) of the OECD TPG (nowhere does Annex I or II say, or imply, “thou shalt now copy vast amounts of the OECD TPG to provide some bulk to your TP document”).
The problem with this information is that it must be read, reviewed and updated year by year; it is huge waste of time, especially where the same TP file, e.g. for a related distributor in Germany, is then also the basis for all other related distributor files across the globe. Worse is that it can give a false sense of security, technical accuracy and completeness.
- Different formats across companies hinders comparability among them. Remember the second objective of TP documentation: providing information necessary to conduct an informed transfer pricing risk assessment? It is not difficult to imagine a scenario where tax auditors may look at 100 companies in their country within a specific industry and pick out 10 for auditing. A first round of sifting will mostly focus on the loss makers. A second round may focus on completeness: can the tax inspector easily find what he is looking for (bearing in mind the layout of Annex II)? If not, those files may be up for further inspection. A third and final round may pick out:
- disorganised files (it is hard to call a file disorganised when it is following Annex II point for point), and
- files with a lack of substance (imagine a 50-page file, with 30 pages from the OECD TPG, 10 pages table of contents, introduction, conclusions and annexes and 10 pages with comparability analyses; such a file lacks substance).
The holy trinity: master file, local file and country-by-country report
For now, I end with the relation between these three documents. In short, these documents together, but also each by themselves, must be coherent.
- In many ways the master file should explain the country-by-country report (“CbCR”): it will be strange if a tax inspector still does not understand the CbCR after having read the master file. In fact, rather than providing extensive explanations of different numbers in the CbCR, one should consider referencing relevant sections of the master file for explanations instead.
- One would also expect any numbers used in the master file to reconcile with, or at least largely agree with, numbers in the CbCR. If this is not the case, it may be prudent to explain that in either or both files (e.g. there are groups which value their subsidiaries at net asset value in the parent companies, leading to an annual multiplication of profits up the ownership chain as each group parent adds its subsidiaries results to each own results for the year in the CbCR).
- Finally, one would expect org charts to match the documentation: e.g.
- the operating entities listed on the master file to match up to the jurisdictions and activities reflected in the CbCR, and
- the functions discribed in the local file functional analyses, to be reflected in the local organisation chart attached to that file:
- it will be weird to find departments on the org chart which are not mentioned anywhere in the functional analysis, or
- to find functions performed in the functional analysis with no place in the org chart to attach them to.
In the next two blogs I will look deeper into master files and local files.