Country-by-country reporting at the OECD – mapping a path to progress

Posted on

I am well aware that many might find my blogging on country-by-country reporting a little repetitive, but when you have been working on an idea for a decade progress towards its achievement is always, I think, worth noting.

Yesterday was important in this respect. At the OECD pubic consultation on transfer pricing documentation it was conceded by the OECD and by all those who might be considered the opponents of CBC that the G8 and G20 mandate to the OECD to introduce CBC meant that it was inevitable that CBC in some form or other would now happen. The important questions now are what is that form and what are the obstacles to progress.

As I made very clear in my presentation to the consultation meeting yesterday (available here), speaking on behalf of the civil society BEPS Monitoring Group of NGO organisations, CBC is a civil society demand and our needs have to be heard and acknowledged when considering the form in which the idea is used. The FT may ignore our role in creating this issue and putting it on the agenda when discussing it today, but the reality is none of this would hve happened without the campaigning of many civil society groups. And what we want from CBC is publicly available accounting data that provides a profit and loss account, limited balance sheet and some cash flow data for each and every jurisdiction in which a multinational corporation works.

As the OECD themselves made clear, that is not what we are going to get out of this consultation. We do, of course, know that. It is not what the OECD has been mandated to deliver, but that does not change our objective. What this current stage in the development of CBC represents is a staging post on the way to what is really needed, which is about so much more than tax reporting and risk assessment, even if we see country-by-country reporting as of immense use for that purpose, as the G8 and G20 obviously do too. That is why it was right to remind the OECD and those at the consultation of this fact: that what is achieved on the way here is but a stepping stone towards a bigger goal. What is more, once it is apparent that this data is available for tax purposes then many objections to it being made public fade away. That is especially true of the cost issue; if CBC data has to be collected for tax (even if quite limited parts of the data collected are reported) the argument that it would be prohibitively expensive to collect it for any other purpose simply disappears because companies will, as a matter of fact, have the information. For campaigners this is a matter of enormous importance. I am also quite sure it explains a great deal of the paranoia on display from big business and its advisers at the meeting on this issue.

It may just be worth summarising - as I did in the brief time available yesterday - just what the information demand we are making is that causes such paranoia. In my slides I said CBC, as civil society wants to see it, provides the following data:

Screen shot 2013-11-13 at 05.39.55

 Deloitte and EY (or Ernst & Young as most will still think of them) led the counter attack. although they were by no means alone (although, in fairness, I should add PWC seemed considerably more realistic). The claims made were numerous.

EY claimed that this data was not available and could not be produced. I was greatly amused by a retort from a business representative who said that any group tax director of any multinational corporation who did not know all this data and the local effective tax rate they paid in every country in which their company worked was not doing their job and should be sacked. I think he put EY in their place more effectively than I could or did: he was right and the meeting knew it.

EY also claimed that disclosing labour data would be commercially dangerous for many companies: this was information that must be kept secret, even from tax authorities. Deloitte had the same paranoia on turnover: they said that to disclose sales by country to tax authorities as an indication of scale of the operation of the multinational corporation in a place was far too risky. Perversely, they preferred labour disclosure for that reason.

There was disagreement on who should prepare the data and supply it. I think on balance the meeting preferred the idea that a head office prepare the data and that it should reflect the group consolidated accounting position. The reason for this has to be understood though, and again it was EY (for a long time the most strident opponents of CBC) who gave the game away as to the reasoning. Their view was that group CBC data was what should be prepared but were emphatic that there should be no requirement to reconcile that data with locally reported accounting in the paces where the multinational corporation traded. I am quite sure that is because the data they want disclosed is that which is included in the group accounts i.e. excluding all intra-group transactions which are precisely where much of the tax risk lies. This would make a mockery of the process, which i am sure is their intent.

They also argued that the data should only be provided by a head office to tax treaty partners i.e. those countries with a double tax agreement with the parent company location. As Will Morris of GE Capital pointed out - quite correctly - this undermined a fundamental purpose of CBC, which is to help with data provision to developing countries, many of whom will not have those double tax agreements. No doubt EY know that too.

Around the room, and despite this, there was a regularly stated concern that the data should only be supplied selectively on a 'need to know' basis ot countries. This is absurd though: existing OECD and UN tax reporting manuals suggest consolidated accounting data should be supplied now to all tax authorities to aid their understanding of the tax risks a MNC's operations in their state create, which is exactly what full CBC data is intended to aid. This argument holds no water as a result; it would fundamentally undermine the reason for CBC data.

So too would the argument that CBC data should only be provided for 'material' states be dismissed as ridiculous: no doubt materiality would be defined in terms of group third party turnover. On this basis firstly intra-group trading would not be a reference point so almost all tax haven activity would be ignored and secondly it is the absence of tax paid that is often of significance (or is, in other words, material) when tax data is considered and such materiality calculations as accountants like to use will, almost invariably, hide such absence from view when scale by size is the accountant's reference point. I regret that, again, I think those suggesting this knew that.

Those suggesting that a 'bottom up' approach to disclosure are also wrong. As has been agreed, CBC is not a basis for tax assessment: it is a risk assessment tool to see whether what a company proposes as the basis for tax assessment in its local tax returns is fair in proportion to the group as a whole. In that case to have CBC data based on local returns is absurd: the two will, for a start, agree and show no discrepancy between group positions and local positions when such discrepancies may well exist at the group accounting level for a number of reasons. Those reasons include the fact that differences in accounting policies, which may be adopted for tax arbitrage purposes, may be revealed by the difference between group and local profits. Or profits created or eliminated on consolidation may well not be disclosed locally but again may have significant tax impact. And, perhaps most importantly, of the group result is disclosed as I suggest, a good indication of the scale of profit due to group activity and so subject to mis-statement through group activity will be available by starting from the consolidated result and working down. There are other reasons too, and try as I might, I can find no reason at all from doing a bottom up approach, not least because more work will be required as this will be a wholly new second consolidation which will not agree with the accounting result, so creating confusion, whilst the outcome will also not show where the group thinks value has really been added, which is vital if the correct allocation of economic profit for tax purposes is to be determined, as far as possible.

This last point is also important. Perhaps the deepest underlying perception of fear I got from those commenting is that CBC is a route on the slippery path to formula apportion taxation. This is explained here by Prof Sol Picciotto, so I will not provide much detail now. Suffice to say that in unitary formula taxation the group profit is the basis for tax assessment and it is allocated to states using a formula that seeks to attribute profit to real economic activity. What are called 'key allocation drivers' are used for this purpose. These drivers reflect the location of those activities that most likely drive profit. Generally those are considered to be sales, labour and tangible assets. What is invariably ignored is intangible assets - for good reason, which I will argue about in a separate blog today, if possible.

Big business hates unitary tax, and for good reason. They defend what is called the arm's length pricing approach (they call it a principle - but it isn't) to transfer pricing as if it is an article of faith. Actually, that is what it is, since it is based on a complete fiction, which is that for every trade there is a market based comparable price that can be determined by comparison with transactions between unrelated commercial entities. This is obviously a fiction because 60% of world trade is intra-group: there's just not enough of the rest to find comparable prices and even if there were the data is rarely available to prove the point. That means that the system of pricing business is dedicated to is fictional - and that lets them weave their own stories on how profit should be allocated. As we all know, the person who tells the big stories often gets away with the world believing them - and in this case that's what big business has done. But it does not make what they do right, rational or even reasonable. Arm's length pricing, as I reminded the meeting yesterday, can work, but clearly does not on a systemic basis or international tax would not be in the crisis it is. In that case defending it as an article of faith - as was argued essential yesterday - is absurd.

Unitary formula tax is based in reality. That is its very obvious merit. It takes 100% of profit and allocates it on the basis of real economic action to states. But that process might - and does - reduce the room for tax abuse. And that is why it is being opposed.

I designed country-by-country reporting to provide the data to make unitary formula apportionment taxation possible. That was not by chance. It was deliberate. Those representing business at the OECD know that. And that's why they oppose CBC, precisely because it will help beat tax abuse.

That's why we need it.

That's why Deloitte and EY are trying to prevent disclosure of data on key allocation drivers.

That's why there is the desire to prevent CBC disclosure on a group basis.

It's also why there is a real wish to deny this information to as many tax authorities as possible as the more that know about it the greater the chance that tax will be appropriately paid.

But CBC is going to happen. The fight now is to make sure it is effective. It will clearly be an issue on which much effort will need to be expended. But it's a fight governments and civil society must win and which big business must lose. That's in the interests of society at large. And it is why the OECD has to act in the interest of society - which ultimately it is meant to represent - on this occasion.

I can hope it will. But if it were more willing to talk to civil society on this issue it would help.


Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:

You can subscribe to this blog's daily email here.

And if you would like to support this blog you can, here: