Are we there yet? Are we there yet?

The Apple state-aid journey rumbles on.  The scene was enlivened somewhat last week with the publication of a white paper by the US Treasury criticising the approach of the European Commission.  The paper dishes out a good kicking and provides a useful template for a company or country considering an appeal to an adverse ruling.

We know most of the key points the US are making.  They are concerned that the US taxpayer could end up footing the bill (as Robert Stack repeats here) but if the tax payments are legitimately due elsewhere then this doesn’t amount to much.  But the risks, functions and assets that generated Apple’s profits were in the US so, under the current system, the tax on those customers is due in the US.  Of course, a company may decide to move those assets but that is an issue that the country of departure has oversight of.  For the period under investigation in the Apple case it is clear that the main drivers of its profitability were controlled and located in the US.

The US Treasury paper looks at the substance of the EC position – that some transfer pricing arrangements put in place (for mainly US MNCs) were “wrong”.  For the EC is this is a competence they do not have nor one that they should be seeking. If you are intent on saying that something is “wrong” you must be able to state what is “right” – but in transfer pricing there are ranges not precise outcomes.

In the case of Apple the EC is looking at transfer pricing arrangements that were put in place in 1991 and revised in 2007.  The cost-plus arrangements put in place for the Irish branches are a fairly standard feature under the OCED’s guidelines for implementing the arm’s length principle.  Applying the OECD’s guidelines would be unlikely to lead to an adverse finding in Apple’s case even if some of the notes and (lack of) documentation could raise concerns.  So the Commission are going to come up with their own guidelines which they have yet to fully reveal.  There is talk of “prudent market operators” and how they would respond to offers made to them to assess the arm’s length principle.

It is a small bit like getting a court summons a couple of years after driving through an 80kph zone at 85kph on the basis that the speed limit is going to be revised down in the future.  Yes, you were over the speed limit but it was probably within an acceptable range so the only way to make a case is to announce a change to the limit but trying to apply that retrospectively is not legally sound.

There have been two significant occasions when the European Commission has told us that the provisions of our corporate tax regime were against the principles of the EU treaties.  The first was against the zero rate of tax for profits from export sales which was abolished in 1980 (with a ten-year grandfather clause) and then against the subsequent dual rate of Corporation Tax of 10 per cent for “manufacturing” and some financial services and, at times, north of 40 per cent for everything else.

These special rates were ruled to be repugnant to the provisions of the EU treaties and were abolished but nobody was silly enough to propose that the rates introduced to replace them be applied retrospectively to previous tax years.  If the European Commission think the national approach to transfer pricing is wrong they might have some grounds for recommending changes but targeting past individual arrangements with a new general approach seems misguided at best.

And remember this is DG Comp not DG Tax. What are they doing getting involved in the minutiae of tax law when they don’t have any expertise, or even significant headcount, in this area?  By all means they should launch state-aid investigations when there is talk of “special two per cent tax rates” but coming up with their own approach to the arm’s length principle, and declaring that to be the de-facto standard, is not a sensible step.

So after much hullabaloo and nonsense about $19 billion the case will come down to the profit attributed to the risks, functions and assets that Apple has in Ireland.  For the main trading company, Apple Sales International, this is the agreement it entered with its Irish branch to carry out the logistics and administration behind the manufacture of its products in China and onward sale to distributors, retailers and some customers.  The taxable income of the Irish branch was determined on a cost-plus basis.  The profit to be taxed was equivalent to 12.5 per cent of the costs incurred by the Irish branch excluding products for resale.  The big sums are in products for resale but these were never owned by the Irish branch.  By 2012 ASI was recording profits of around $35 billion.  The publicly available information on ASI is collated here.

The second company under investigation, Apple Operations Europe, has a branch in Ireland that carries out the manufacture of a specialised line of computers and mainframes.  It is the only manufacturing facility that Apple owns.  The amounts here are relatively small and any adjustments to the transfer pricing arrangement are unlikely to be significant.  AOE does own 99.99 per cent of ASI but as both are non-resident it is the profits of their Irish operations that matters not their overall profits.  Both companies were established in 1980.

Any significant sum is likely to come from proposed changes to the cost-plus arrangement for determining the taxable incomes of Apple Sales International’s Irish branch. When looking at the 2011 figures the Commission have indicated a return–on-sales margin of 0.2 per cent would result in the same taxable income as given by the cost-plus margin of 12.5 per cent.  A return-on-sales margin of one per cent would give an additional tax bill of around €200 million for the ten years in question.  Upping the margin to three per cent gives a tax bill of around €600 million.  How much would a “prudent economic operator” need to be paid to handle the logistics and administration behind Apple’s manufacturing and distribution?  When the transfer-pricing arrangement was revised in 2007 the total sales of ASI were around €6 billion.  Getting three per cent of that would equate to a profit of around €180 million.  Where do I sign??

Cost-plus arrangements are common but there are alternatives.  Here is an extract from the 2006 accounts of Google UK Ltd:

Turnover represents the total intercompany cost plus 8% charged to Google Ireland Limited for sales and marketing and cost plus 10% for research and development charged to Google Inc. during the year, excluding value added tax.

Earlier this year HMRC – who are tax experts – completed a long audit of Google.  The outcomes for the 10 years under audit was that Google had to pay £130 million in additional UK corporation tax – with most of this apparently related to how share-based remuneration is treated in the calculations of Google UK’s taxable income.

It also seems that from now on Google UK’s taxable income will be a function of the size of the market it services (return on sales) rather than the size of its operations (cost plus).  At the conclusion of the audit Google said:

“We have agreed with HMRC a new approach for our U.K. taxes and will pay 130 million pounds, covering taxes since 2005. We will now pay tax based on revenue from U.K.-based advertisers, which reflects the size and scope of our U.K. business.”

We don’t know the return-on-sales margin to be applied to this new  approach but it does show that cost-plus and return-on-sales approaches can potentially be interchanged.  HMRC did not, though, require that the return-on-sales approach be applied retrospectively.

So what are the European Commission going to do in the case of Apple? We don’t know. It is only a guess but it is possible they will propose that the taxable income of ASI’s Irish branch should have been assessed as a margin of the sales that it provided support for.  The resultant tax bill depends on the actual margin proposed but it will not be $19 billion.