Old chestnuts roasting in the public domain

Eaton A.P.A. cancellations were an abuse of I.R.S. discretion

This article appears in Insights vol. 4, Issue 9.  Insights is the Tax Journal of Ruchelman PLLC.  As the transfer pricing travails of Eaton Corporation (“Eaton”) continue, a recent Tax Court decision affirmed that I.R.S. administrative procedure set down in Revenue Procedures and relied upon by the I.R.S. and a taxpayer cannot be arbitrarily circumvented, and that the I.R.S. must reasonably exercise its discretion.

At issue was the cancellation of two advance pricing agreements (A.P.A.’s) and the consequent I.R.S. income adjustments made as a result of applying a new transfer pricing method.  Eaton’s position was that the A.P.A.’s were binding contracts, and that these contracts were cancelled for reasons other than those named as cause for termination in the respective A.P.A. agreements.  Though the Tax Court did not agree with Eaton that an A.P.A. agreement should be interpreted under contract law, the Tax Court carefully reviewed the circumstances of the cancellation against the I.R.S. Revenue Procedures Rev. Proc. 96-53 and Rev. Proc. 2004-40 that governed the drafting and administration if the A.P.A. agreements in the relevant tax years.

An A.P.A. is an alternative to the traditional adversarial model between a taxpayer and one or more tax authorities.  Its purpose is to reach an agreement concerning the transfer pricing method to be used for a number of tax years in one or more controlled transactions.  A.P.A.’s take a long time to negotiate, owing both to the fact-intensive nature of transfer pricing matters and the considerable due diligence both sides must undertake.  Both sides must be prepared to compromise their technical positions somewhat in order to obtain practical transfer pricing certainty.  Once concluded, an A.P.A. agreement is signed and a program of annual reviews undertaken to ensure that the terms of the A.P.A. are being followed.  For transfer pricing positions that influence a large share of a tax provision, or a significant transfer pricing position that is complex or unique, generally accepted convention holds that is it better to spend the time and fees for two years negotiating an A.P.A. than to spend an even greater amount to produce I.R.C. § 6662 documentation, manage examinations, Appeals, Competent Authority and litigation.  Currently, the administrative procedures for requesting and administering an A.P.A. are set out in Rev. Proc. 2015-41.

As with any agreement, the hallmarks of a successful A.P.A. are negotiation in good faith, disclosure of all material or relevant facts or documents, disclosure of true facts or documents, and the adherence to the terms of the agreement over the duration of the agreement’s lifespan.  The I.R.S. alleged “failure of a critical assumption, misrepresentation, mistake as to a material fact, failure to state a material fact, failure to file a timely annual report, or lack of good faith compliance with the terms and conditions of the A.P.A.”[1] as among the “numerous reasons” for the cancellation of the Eaton A.P.A.’s.  The Tax Court weighed each claim, finding in favor of Eaton in the case of all stated reasons for cancellation.

The origin of the dispute was a series of inadvertent accounting errors committed by Eaton accounting and tax personnel, and discovered only after a new transfer pricing manager joined the company and looked de novo at the calculations and underlying information used to comply with the terms of the A.P.A.’s.  Many of these errors did not result in a favorable tax outcome for Eaton, though in net terms the transfer prices were higher as a result by approximately 5% in each of the 2005 and 2006 tax years.  Eaton alerted the I.R.S. to these discrepancies, filed Forms 1120X to report the additional income, and prepared to update its annual A.P.A. reports to explain the effect of the errors.

In response, the I.R.S. changed its view concerning the transfer pricing method in negotiations of a third A.P.A., advised Eaton not to file updated A.P.A. reports, and issued a letter cancelling the A.P.A.’s covering tax years 2001-2009.

The extensive information gathering and questioning that occurred during the first and second A.P.A. negotiations, as well Eaton’s responsiveness and cooperation, proved to be a large part of the undoing of the I.R.S. case.  Many items of information that were alleged to have been omitted or neglected by Eaton in an act of bad faith bargaining were found to have been disclosed during A.P.A. negotiations, or alternatively could have been discovered by the I.R.S. during its many series of questions or meetings.

In the end, the conditions for the cancellation of an A.P.A. set out in the Revenue Procedures were not met, as the Tax Court found in its analysis that concentrated inter alia on the interpretation of the terms “material fact”, “critical assumption” and “misrepresentation” in the context of the Eaton facts.  It was noted that either side could have walked away during either of the two A.P.A. negotiations if viewpoints concerning the best transfer pricing method differed significantly enough, and that the I.R.S. signing of two largely similar A.P.A. agreements limited its ability to argue in retrospect for a different transfer pricing method.

The Eaton case outcome highlights the complexities of implementing a transfer pricing method once the “transfer pricing study” is complete, especially when accounting and enterprise information systems are used to store information and generate reports to be used in tax calculations.

The 202 page Tax Court memo explains in considerable detail the data warehousing procedures used by Eaton to store report templates and files, the ledgers and “mirror ledgers” used to record transfer prices for accounting purposes and eliminate intercompany transactions on consolidation, and the sources of information used to calculate ratios and cost variance factors critical to the compliance with the A.P.A. terms.  It was the fact that many data sources had to connect across group companies using the intervention of controllers and tax personnel that informed the finding of “human error” or “computational error” rather than deliberate misrepresentation or deceit.  Even with an audit opinion on the non-consolidated financial statements of the Eaton entities relevant to the A.P.A.’s, the calculation errors slipped by the taxpayer and the I.R.S. at successive annual reporting checkpoints.

While a relief for companies with pending or in-force A.P.A.’s or Competent Authority settlements, this decision illustrates the value of proper transfer pricing policy implementation and the engagement of employees and advisors outside of the tax function to make sure the system works reliably from the start.  People entrusted with key information or process control can change jobs unexpectedly or eventually retire.  Initial engagement of all people needed to produce results representing true taxable income is critical, as is a periodic check to ensure that the system is performing as expected.

Finally, Eaton serves as a reminder of the possible unfortunate consequences of fixing mistakes, even honest mistakes, in a climate of heightened suspicion of tax avoidance among tax authorities.

[1] Eaton Corporation and Subsidiaries v. Commr. T.C. Memo 2017-147, p. 112.

Amazon makes the CUT – an important taxpayer win, a reminder to consider transactional evidence

This article also appears in the May issue of Insights, the Tax Journal of Ruchelman PLLC

In finding for the taxpayer in a recent transfer pricing decision,[1] the U.S. Tax Court followed its own determination in Veritas[2] in valuing a buy-in payment made as compensation for the right to use pre-existing intangible property in a related-party cost-sharing arrangement (“CSA”).  This decision, like other major transfer pricing decisions, serves as a reminder of the fact-intensive nature of transfer pricing matters and of the importance of uncovering and properly analyzing transactional evidence from the controlled transaction in question and from uncontrolled transactions or dealings of the business.

Amazon.com, Inc. (“Amazon”) entered into a CSA with Luxembourg subsidiary AEHT in 2005.  The CSA covered: (i) the software and other technology underlying the Amazon European domain-name websites, fulfillment centers, and related business activities; (ii) marketing intangibles, including trademarks, tradenames, and domain names used in Amazon’s European business; and (iii) customer lists, customer data, and the Amazon tradename and mark.  The right to use the pre-existing intangible assets in these three categories was priced at $254.5 million, payable over a seven-year period corresponding with the useful life of the intangible assets.

Using the income method and the same discounted cash flow approach rejected by the court in Veritas, the IRS estimated the arm’s length value of the buy-in payment to be $3.468 billion, effectively disregarding the CSA and valuing the transfer of rights as a business that would exploit infinitely-lived intangibles in perpetuity.  The IRS also disputed the Amazon failure to classify certain technology and content expenses of Amazon.com as intangible development costs, thereby biasing downward the income from annual cost sharing payments received from AEHT over the term of the CSA.

The economic substance of the AEHT Luxembourg operations hub was not critical on its own.  Local language requirements, local vendor relations, and European logistics considerations and customer tastes were all factors contributing to the need to carry on a business in Luxembourg, and the change in economic position for AEHT expected to result from the CSA  In rejecting the IRS transfer pricing method, the court made clear that “AEHT was not an empty cash box.”  This determination contrasts sharply with the O.E.C.D. outcomes under the BEPS Action Plan that attack hypothetical “cash boxes” that are legal owners of rights but lack the decision-making and risk management capacity needed to allocate capital to investments with uncertain returns.  The dispute in Amazon therefore centered on (i) the transfer pricing method, (ii) the assumptions made and analyses used to value the buy-in payment, and (iii) the correct treatment of the intangible development costs within the term of the CSA

In deciding for the taxpayer, the court relied on the testimony and reports of 30 experts in computer science, marketing, economics, and transfer pricing economics.  The opinions of the computer science experts on the state and viability of the Amazon software and websites served as a stable foundation upon which the transfer pricing economics experts for the taxpayer could anchor their assumptions.  These assumptions were critical – as the technical constraints of the software system provided a reliable estimate of the lifespan of the software used to power the core operations of the Amazon websites and fulfillment business.  The marketing experts helped the court decide on a proper method to estimate key variable values used in the marketing intangibles value calculation.  They also assisted the court in determining how the intangibles allowed a team of engineers – for whom no technical challenge seemed too large – to overhaul the websites without causing them to crash during popular shopping seasons.

However, the star of the trial was a Treas. Reg. §1.482-7 transfer pricing method – the Comparable Uncontrolled Transaction method (“CUT”).  Amazon used an unspecified transfer pricing method resembling in some respects the profit-split method to calculate the original buy-in payment, while the IRS used an application of the income method.  The IRS income method calculated the present value of cash flows forecasted to result from AEHT’s European business, using cash flow and balance sheet forecasts as its only company data input.  Both approaches neglected or devalued Amazon’s outsourced web store programs, and thousands of Associates or Syndicated Stores programs that provided customer referrals to Amazon.

The website platform and referrals transactional data alone did not win the case for Amazon.  Considerable expert testimony was required to establish reliable assumptions of discount rates, value decay rates, useful asset life, and trademark ownership.  The company’s own information, however, was a crucial element in winning the case.  CUT’s that involve transactions between the taxpayer and independent businesses (sometimes called internal CUT’s), are highly persuasive given these fit well within the framework of the comparability requirements of Treas. Reg. §1.482(c)(1), which is critical to selecting a best method.  CUT’s are not abstract agreements between third parties.  They must bear some resemblance to one of the controlled parties and its business.

One small levity allowed in the 207-page decision was that “one does not need a Ph.D. in economics to appreciate the essential similarity between the DCF methodology that Dr. Hatch employed in Veritas and the DCF methodology that Dr. Frisch employed here.”  Similarly, a Ph.D. is not required to present a well-selected and adjusted CUT to the IRS or a Tax Court judge.  It seems unlikely in the case of Amazon’s CSA that the IRS would have paid any attention to a CUT at the examination level, given the strong motivation within the IRS to re-litigate Veritas.  Nonetheless, CUT’s remain a valuable commodity to be mined and stockpiled for use in appropriate circumstances.

Not only was Amazon’s transactional data important in building its case in favor of the buy-in payment value, its Code §41 credit cost detail proved useful in substantiating the company’s claim that a significant class of expenses should not be classified as intangible development costs and shared with other CSA participants.  This is another good example of seeking the data required within the company’s records before reinventing the wheel.

An open question in the case is the treatment of employee stock option costs in a CSA.  This question will have to wait for the outcome of the IRS appeal in Altera.[3]

Pending a successful outcome in Altera, two theories used by the IRS to attack a technology company CSA could be blunted.  To the extent that IRS estimates regarding the size of the tax gap rely on large income windfalls from litigating CSA positions of high-tech companies, Amazon could prove to be an early indication that these estimates require a downward adjustment.

[1] Amazon.com, Inc. & Subsidiaries v. Commr., T.C., 148 T.C. No. 8 Docket No. 31197-12.

[2] Veritas Software Corp. v. Commr., T.C., 133 T.C. 297 (2009).

[3] Altera Corp. v. Commr., T.C. 145 T.C. No. 3 (2015).

Is the CRA skipping dinner in the hope of getting its deserts? Digesting the “new” Canadian transfer pricing documentation standard

The CRA was recently asked “Will the CRA’s expectations of the “reasonable efforts” that a taxpayer must make to determine and use arm’s length transfer prices include the preparation of transfer pricing documentation that is consistent with the recommendations from the OECD in Action 13 of the BEPS initiative (i.e. Master File and Local File transfer pricing documentation)?”

This is a reasonable question to ask in view of the years of work of the Canadian government delegates to the OECD Base Erosion and Profit Shifting Project (BEPS), and the endorsement of the final BEPS Project Action 13 Report by the Government of Canada in the latest federal budget.  Perhaps some were expecting harmonization with the new international norm, or more precise guidance on how to avoid penalties than can be found in the current six bullet points in paragraph 247(4)(a) of the Income Tax Act and a smattering of guidance.  The response suggests something different altogether.

Answer: “The CRA considers that BEPS Action Item 13 has been dealt with by the introduction of proposed section 233.8 of the Act relating to Country-by-Country Reporting. The “reasonable efforts” requirement is based on the legislation contained in section 247 of the Act, in particular the requirement to produce “contemporaneous documentation” in accordance with subsection 247(4). Proposed section 233.8 has no direct relation to section 247 and does not include a specific requirement to produce a “local file” or “master file”. As such, the CRA has not altered its criteria regarding whether a taxpayer has made reasonable efforts to determine and use arm’s length transfer prices.”

Like a quietly disappointed restaurant patron wishing to be left alone to finish her mediocre dish, the CRA was in effect asked how it was enjoying its meal of the OECD’s new transfer pricing documentation guidelines.  Its response suggests local file and master file requirements were too much of an acquired taste, and the CRA has now sent those items back to the kitchen.  The diner’s overall response of “Fine…” derives only from the great pleasure the Agency may expect from a dessert of Country-by-Country reports, and from being able to claim the OECD confection has been adopted and was a good choice for Canada.

The prospect of the empty calories from the Country-by-Country data tables must be too thrilling to risk filling the stomach with the main course, as many other tax authorities are learning to do.  (CbC reports, like desserts, contain relatively little nutritional value, and are only intended to look at for risk assessment purposes, not to make a meal of for easy reassessment calculations.)  A feast of taxpayer data and a maintained state of uncertainty keeps the playing field tilted in favour of the CRA.

Many foreign tax authorities have adopted the Action 13 report as a whole as well as other new transfer pricing principles, in line with the intentions of their finance ministers.  Many tax administrations have also taken on board the updated OECD Transfer Pricing Guidelines.  These tax authorities will require the master file from foreign subsidiaries of Canadian-parented companies as part of their documentation requirement.  It will be burdensome for Canadian taxpayers to have to produce more documentation than is ostensibly needed (and risk creating more than one set of facts when double tax cases arise).  An opportunity to mitigate the median taxpayer’s burden has clearly been passed up.  The “new” documentation burden will fall disproportionately on Canadian companies with worldwide net sales under $ 1 billion (approximately €750M at today’s exchange rate) that do not have to file a CbC report, but also do not have extensive tax departments.  These mid-sized and smaller companies must now contend with the compliance cost of the CRA double standard.

If the CRA can ignore adding parts of the BEPS Project transfer pricing reports  to Canadian transfer pricing rules (despite those additions being endorsed by the most recent federal budget), one might ask what will be ignored next.  The CRA has stated publicly that it already takes similar positions to those set out in writing in the various BEPS transfer pricing final reports, making it seem the international guidance is of diminished value in Canadian transfer pricing matters.  Will the CRA adopt the OECD Guidelines as updated by the BEPS project, or stick with the 2010 version it finally recognized in October 2012?  The historical approach suggests it is appropriate to avoid setting down the CRA position in writing for some time.  Perhaps the CRA is waiting for the U.S. Department of the Treasury to walk away from the BEPS Project consensus after Donald Trump enters office, is hedging its bets in anticipation of several Tax Court decisions, or is really more interested in penalties and reassessments than encouraging compliance.

Taxpayers that have wondered “How good is good enough?” when it comes to interpreting the Canadian documentation requirement will continue to be offered a diet of junk food by the CRA and asked to keep guessing.  Unfortunately for taxpayers, the real rules are off the menu and appear only when requested directly by professional tax advisors.  Taxpayers owe the Canadian Tax Foundation a debt of gratitude for this revelation.

Canadians are known the world over for many great traits, but in taxation are increasingly known for picking the luke-warm bowl of porridge.  The made-in-Canada approach (or compromise) is also a favourite.  Companies will need to carefully consider how to approach this particular tepid bowl of Canadian porridge.

A new way to do the splits: BEPS draft guidance falls short of enabling global formulary apportionment

It takes considerable training and the right physical conditioning to successfully do the splits and avoid injury or embarrassment.  For those who view transfer pricing as a gymnastics sub-discipline, applying a profit split method is often an approach of last resort and is arguably as difficult to accomplish in a graceful manner as the gymnastic feat.  Since the BEPS (Base Erosion and Profit Shifting) Project began in 2013, a key focus the revision of the OECD guidance that multinational companies and tax authorities use to apportion income resulting from the joint commercialization of intangible assets within a multinational group.  The unwelcome, and potentially widespread, ex-post use of profit split methods as proxy for global formulary apportionment was viewed by corporate taxpayers and commentators with the same sense of dread as a surprise gymnastics skills test.

However, it would appear that companies can relax somewhat after the July 4 publication of the OECD Revised Guidance on Profit Splits discussion draft.  The discussion draft links other transfer pricing developments in the BEPS Project[1] to the guidance on the application of the transactional profit split method, but it does not propose to place an over-broad profit apportionment tool in the hands of tax authorities.

Like a gymnastic maneuver, successful application of the transactional profit split method requires a full understanding of risk – in this case economically significant risk incurred by the participants in the relevant business opportunity.  The transactional profit split method is one of five transfer pricing methodologies set out in Chapter II of the OECD Guidelines.  In cases where controlled taxpayers participate in highly integrated operations and contribute valuable intangible assets in respect of a joint business opportunity, the profit split method is used to split the profits or losses from the combined activity on an economically valid basis to approximate an arm’s length return to the respective contributors.

Not unlike the provisions of Treas. Reg. §1.482-6, the transactional profit split may be applied using either the more direct contribution analysis or the more indirect residual analysis (i.e., routine profits to the associated enterprises are determined first, and then deducted from the actual pooled profit to determine the residual profit to split).  The transactional profit split method can also be used in conjunction with a valuation method to estimate the value of an intangible asset transferred from one controlled taxpayer to another.

In contrast to the Treas. Reg. §1.482-6 method, the OECD Guidelines allow for the splitting of either anticipated or actual profit.  The discussion draft adapts the OECD Guidelines profit split by incorporating the changes to Section D.2.6.2 of Chapter VI that discuss how to reliably estimate anticipated profit from an intangible asset.  The draft properly points out that appropriate use of the transactional profit split method uses a profit split metric determined in advance of the knowledge of the actual profit to be divided between the two parties.  This serves as a reminder to companies of the evidentiary value of intercompany agreements – used in this instance to demonstrate taxpayer intent and to clearly set out the way in which a split of unanticipated profit will be calculated in the future.  The fact that an agreement is required to manage the uncertain outcome of a business activity where risk is shared, in and of itself, reinforces the appropriateness of a profit split method.

The use of the transactional profit split method based on the combined actual profits of the contributing parties is linked to the control exercised by those parties over the economically significant risks associated with the combined business.  The transactional profit split method may, therefore, not be appropriate in circumstances where the risks of the combined business are not separately or collectively controlled by the participants, or where each party does not have the financial capacity to assume its proportional share of the risk.  The evaluation of control over risk should be carried out annually, as actual profits are intended to be split each taxation year under the transactional profit split method.

This limiting control condition arises from the work completed by the BEPS Project, to date, on transfer pricing issues relating to intangible assets.  Interestingly, this new limitation on attribution of profit from intangible assets to only those entities exercising control over risk and possessing sufficient financial resources to mitigate risk circumscribes the authority of tax administrations to use the transactional profit split method in a formulary way, as was feared by many BEPS Project observers.

Some useful guidance appears in the discussion draft to differentiate a reliable profit split from a less graceful version.  Parties must “share the same economically significant risks”[2] associated with the combined business activity or “separately assume closely related risks”[3] associated with the same activity.

The term “economically significant risks” is explained in the revised Chapter I of the OECD Guidelines[4] as being those factors that cause the anticipated objectives or outcomes of the business activities for the contributing parties to vary to the greatest degree.  Strategic risks, marketplace risks, infrastructure risks, operational risks, financial risks, transactional risks, and hazard risks are suggested as the principal (though not the only) types of risk to consider.

There is, therefore, a less reliable profit split where

• the economically significant risks have not been specified,
• the nature of the contributions of the parties has not been accurately determined,
• an evaluation of how those contributions influence profit outcomes has not been made,
• the profits to be split have not been reliably identified, and
• the basis for splitting the profits has not been reliably determined.

In certain cases, tax authorities (and sometimes companies) choose to skip the difficult work of comparability analysis or comparability adjustments, and apply the profit split method.  The discussion draft acknowledges a shortage of comparables may exist in practice, but it warns that a lack of comparables alone does not justify the use of the transactional profit split method.  Rather than stretching to apply the transactional profit split method, the discussion draft suggests that the use of a different method (inexact comparables) and well-supported comparability adjustments may result in a pricing outcome that better approximates an arm’s length result.

Similarly, the discussion draft sets out limitations, concerning integrated operations, unique and valuable contributions of intangible assets, and group synergies to the use of the transactional profit split method, in order to promote the responsible use of this transfer pricing method.  The mere appearance of integrated operations is stated as an insufficient condition for the application of the profit split method.  A careful functional analysis and an understanding of the company’s value chain is required to establish whether it is truly the case that the functions of company participants are so integrated that an intercompany transaction cannot be reliably delineated and perhaps priced using a more reliable methodology.

Finally, the discussion draft clarifies that treatment accorded to profits resulting from group synergies should differ from the treatment of profit resulting from the commercialization of intangible assets.  The benefits or cost savings connected with group synergies are termed “marginal system profits,” which should not be included in the “total system profits” to be divided using the transactional profit split method.

Room for disagreement exists with regard to the definition of a unique or valuable intangible asset, the degree to which a risk is economically significant, the importance of location savings, and the way market characteristics figure into a profit split analysis between a company based in a country with a developed economy and a related party with operations in a country with an emerging economy.  Nonetheless, the focus of the OECD guidance on intangible assets has been sharpened significantly, thereby reducing uncertainty for all.

[1] See, e.g., OECD, Aligning Transfer Pricing Outcomes with Value Creation, Actions 8-10 Final Reports, (OECD Publishing, Paris: 2015) (the “OECD Guidelines”).

[2] OECD, Public discussion draft, BEPS Actions 8 – 10, Revised Guidance on Profit Splits, (OECD Publishing, Paris: 2016), para. 16.

[3] Id.

[4] Supra note 1, Section D.1.2.1.1, pp. 25-28.

Mind the $2 billion gap – Medtronic decision for the taxpayer favors pricing transactions over profit split

The U.S. Tax Court recently decided for the taxpayer in a $2 billion transfer pricing adjustment imposed by the IRS [Medtronic, Inc. v. Commissioner, T.C., No. 6944-11, T.C. Memo No. 2016-112].  At issue was the pricing of two licenses – one in respect of medical device technology, and one covering the Medtronic names and marks – and which transfer pricing method is the best method to apply under Treasury Regulations § 1.482-4.

Medtronic took the approach of looking to uncontrolled licensing transactions for support in applying the Comparable Uncontrolled Transaction (CUT) method, while the IRS used the Residual Profit Split Method (PSM).  Broadly speaking, the CUT prices the transaction directly in a manner that follows how most third parties make IP deals.  The PSM, on the other hand, splits the system or joint profit left over after allocating a basic return to the contributions of the transaction participants.  The art in the PSM is deciding how most reliably to split the profit that’s left over – a value that can be large where valuable intangible assets are used in a business at scale.  Reasonable people can disagree over works of art.

The difference between the outcomes of two approaches was colossal (no, that 59% royalty rate was not a typo), even after the dispute had been through the IRS audit process for some time.  To an experienced transfer pricing economist, a large difference in the results obtained from more than one transfer pricing method means only one thing.  An error or errors, a questionable assumption, a data problem, or incomplete information.  In this instance, the experts for the petitioner and the respondent were not of much help to the judge in exposing the reasons for the gap between the two positions.  The judge was left to go back to the complex facts and make her own decision – as has been the case in other transfer pricing trials.

Without spending considerable time determining whether the taxpayer’s approach was reasonable, the decision dismantles the IRS PSM analysis thoroughly, relying on the facts rather than the assumed consequences of company characterization clichés –  “contract manufacturer”, “limited risk”, “routine intangibles”, etc.

The decision sounds a warning for those currently parroting the undefined OECD term “value creation” as a determinant of the location of company profit.  The court found the IRS “value chain” analysis to have no standing as a transfer pricing method, and relied on a misinterpretation the business facts to fit the theory of the PSM approach.  Alternatively, the decision evaluates what attributes made Medtronic’s products worth purchasing and paying for (and also less worth purchasing and paying for) to its physician and patient customers.  A causal connection was found between market share and market pricing, product quality outcomes, and the companies, departments and employees that contributed to an effective and safe product, final assembly and product sterilization, final quality checks.  The decision also carefully identified which companies incurred losses when implanted products failed or were recalled, and used this information in its comparability analysis.

Underlining the importance of good intercompany agreements, the decision relied on the intercompany contracts the parties had concluded to explain their responsibilities and the division of liability in the event of a quality problem.  The contracts could be relied on, as the contractual terms matched the actions of the parties.

This taxpayer win, which may yet be appealed, is instructive for companies that must perform a best method analysis under US tax law, encouraging to companies now experiencing an unexplained IRS breach of an agreement on the method of determining true taxable income, and foreshadows the release of controversial OECD guidance on the application of profit split transfer pricing methods.

Canadian transfer pricing rules – scattered from far and wide

The recent March 22 budget announced an update of the transfer pricing guidance used by the CRA.  The main change was the adoption of the revised OECD Transfer Pricing Guidelines, last amended in October 2015 as a result of the OECD/G20 BEPS Project.  While this guidance does not have the force of law, it is used by the CRA in transfer pricing audits as well as selectively by Canadian courts.

This announcement itself is noteworthy only in its timing, given CRA has deviated somewhat from the international consensus on the implementation of the BEPS Project results.  More notable is that Canadian transfer pricing guidance is now scattered across 22 documents – add a few more for large companies required to file a Country-by-Country report.  This count does not include the law or transfer pricing decisions from the courts, which are critical to consider.

The result will be awkward and time-consuming reading until the OECD Guidelines are properly reissued in the next few years after the conclusion of further work.

In the long-run, taxpayers and their advisors should ideally be able to reference two or three documents, in addition to relevant jurisprudence.  We’ll update you once we get there.

Is the CUP half empty?

In its recent decision in Marzen Artistic Aluminum Ltd. [2016 FCA 34], the Federal Court of Appeal upheld the Tax Court’s use of a contract between two independent parties as a Comparable Uncontrolled Price.[1]  One of those independent parties was a participant in the related-party transaction whose pricing was at issue, and the other was a management company.  Is Marzen Aluminum the new exception in the rationale for the use of the CUP method, or the new rule (in Canada, at least)?

Suppose Canadian corporation A is related to non-resident corporation B, and A contracts with B to perform a single type of service.  B then contracts with C, an independent non-resident individual or corporation to provide substantially all the same types of services that B is obligated to provide to A.  B can meet its obligation to A by hiring no further resources other than C.

Should the price charged by C be used to set the transfer price charged by B?  Though both the Tax Court and the Federal Court suggest this approach makes sense, the facts of Marzen Aluminum were unique, meaning this approach might not always apply.  Comparability standards set out in the OECD Transfer Pricing Guidelines require that the correct answer to this question should address both the facts of the transaction and the economic circumstances of the transacting parties.

The utility of the Marzen Aluminum decisions is limited for two main reasons.  First, as different services were rendered by B and C involving fundamentally different risks, any comparability analysis is arguably of limited future value.  Second, the ranking of transfer pricing methodologies that once held the CUP method to be the most reliable has been replaced by the more objective ‘most appropriate method’ approach introduced in the 2010 version of the OECD Guidelines.  The standard of comparability required to use the CUP reliably has increased between 1995 and 2010.

Marzen Aluminum may suggest that pricing a related-party transaction using the CUP method now leads to greater uncertainty, or that the CRA can now successfully rely on a bygone standard.  If accepted comparability standards and analytical rigour are employed, the CUP will remain half-full.  Using another transfer pricing method?  It may runneth over the CUP.

[1] CUP – one of the five transfer pricing methods accepted by the Canada Revenue Agency and other tax authorities

New transfer pricing guidance from the OECD – A Hitchhiker’s Transfer Pricing Guidelines to the Galaxy

The October 5, 2015 releases from the controversial OECD/G20 BEPS Project are finally here.  While there is still work to do on a number of the transfer pricing matters, companies now know what will be considered by tax authorities in modernizing country transfer pricing rules and guidelines.

The outcomes of the BEPS project, on one hand, mean nothing quite yet.  The OECD has issued tax policy and guidelines.  For these guidelines to be meaningful there must be formal adoption under the law.  The G20 members and other BEPS Project participants are committed to proceeding from policy to implementation.  An appropriate reaction might follow Ford Prefect’s mantra: “Don’t panic”. 

Companies can be fairly certain that the new guidance will be followed reasonably closely by G20 member states and others.  It’s therefore good to be prepared like Arthur Dent (who always knows where his towel is) for an eventual meeting with the Vogons of the world’s tax administrations under the amended OECD Transfer Pricing Guidelines.  After all, a number of countries have already started enacting BEPS-inspired legislation, and adopting BEPS-like administrative positions.  Here is what the future looks like, in brief, and from a high altitude:

• A related party transaction will be priced given what it is (or can be proven to be) in fact, not what it says it is on paper. Contracts will be respected if their terms are the same as actual expectations or outcomes.  Well drafted contracts are now essential, not optional.
• Risk connected with a related-party transaction can be moved or reassigned by tax authorities between related companies where an entity does not demonstrate control over risk. This will mean having appropriate people making investment decisions and managing and controlling the resulting risk over the lifespan of the investment.
• Capital at risk without control over risk occurring in the entity that invests the capital, with only limited exceptions, will attract only a risk-adjusted return.
• Intangible assets are now defined more broadly. Location savings and market characteristics, argued by China and India to be intangible assets, are now important comparability factors to consider as determinants of price or profit.  One-sided approaches that give all intangible returns to one or few group companies will have to be reviewed under the new guidance.
• Valuations of intangible assets and contributions of intangible assets to cost sharing arrangements will receive considerably more scrutiny, and will be revised in certain circumstances after the transaction has occurred if forecasts and actuals differ by more than 20%.
• The arm’s length principle will continue to apply. However, more work is being done on profit split techniques at the OECD.  Companies will have to present more robust transfer pricing analyses using an OECD method to convince a tax authority that the profit split is not the most appropriate method to apply.
• Transfer pricing documentation requirements will become more onerous and detailed in most countries, and a ‘master file’ describing the worldwide business will need to be produced and kept up to date and on file.
• Country-by-country reporting requirements begin to take effect in 2016 for large corporations with global revenues exceeding €750 million. Tax authorities will have more data to use in assessing transfer pricing and other risk, and will have to guarantee data security and responsible use.
• Different countries will read and interpret the 186 pages of guidance and examples issued on October 5 differently. Just as transfer pricing disputes between tax authorities and taxpayers is expected to increase, so too will the resulting number of double taxation cases to be resolved between countries.  Tax authorities are generally not well prepared to handle the expected significant increase in cases.  Expect double tax to persist for a longer period of time.  The OECD continues work on dispute resolution procedures and tools.

In the real Hitchhiker’s, a computer called Deep Thought is designed to calculate the “Answer to the Ultimate Question of Life, The Universe, and Everything”.  After 7.5 milion years, the answer that resulted was 42 (it should have been 54).  The OECD took 2 years to answer a slightly less difficult question about international tax.  Don’t panic.

Making popcorn from a kernel of truth – Does the U.S. Tax Court decision in Altera have broader consequences?

The U.S. Tax Court’s recent decision in Altera v. Comr. affirmed that that the arm’s length standard controls over another specific regulatory provision, in this case the commensurate-with-income standard of Treas. Reg. 367(d).  The decision in Altera found that stock-based compensation is not an expense that should be pooled and shared under a qualifying cost sharing arrangement.

Altera’s support for its position was largely empirical, and stood on the shoulders of the case built in Xilinx for exclusion of stock-based compensation from a cost pool in a cost sharing arrangement.  This empirical evidence consisted of a number of joint venture and other collaboration agreements submitted by commentators to the 2003 proposed cost sharing regulations.  The Tax Court was persuaded that not only is stock-based compensation not shared between cooperating independent parties; it held that stock-option expense should not be considered as an element of the comprehensive set of costs considered by Treasury to be “relevant costs” in a qualifying cost sharing arrangement.

These agreements included certain elements of labor compensation that parties to the agreements consented to share, but did not include stock-based compensation among those expenses.  Agreements from the software industry, comparable to the industry in which Altera operated during the years at issue were produced and viewed as sufficiently comparable to the Altera arrangement at issue. These agreements proved persuasive in Altera, and served to amplify the effect of the failure of Treasury to consider the submissions from commentators to the 2003 proposed cost sharing regulations.

Neither Treasury, as part of its finalization of the 2003 regulations, nor the IRS in Altera produced evidence of an agreement between third parties that included stock option cost.  Proof of arm’s length behaviour with respect to stock option expense was therefore delivered in the form of a negative empirical result.

To rebut the arguments advanced by the taxpayer, the IRS relied on the commensurate-with-income standard, and did not present any expert opinion that supported its position that stock-based compensation must be included in the cost pool of a qualifying cost sharing arrangement to achieve an arm’s-length result.

If the U.S. Tax Court can strike down regulations that are not consistent with the arm’s length standard, one might think about the prospects for other aspects of Treasury Regs. Section 482 that are not based entirely on empirically supported independent corporate conduct, or consistent with the definition of the arm’s length standard in Section 482-1(b)(1).  Some areas to consider might be:

• The empirical support for the factor of 1.3 used in Section 482-2(a)(2)(iii)(B)(1) to set arm’s length interest rates based on the AFR, or the use of the AFR more generally
• The relevance of the lists of specified covered services, excluded activities, or the concept of low-margin covered services in the Section 482-9 services regulations when compared with documented business practices concerning the price of certain services types
• Joint venture arrangements or other transaction evidence used as comparables for intangible assets transactions that do not depend on a valuation technique akin to the income method

Oddly, this self-criticism may be good for the U.S. Treasury Department at a time of threat to the arm’s length standard from the G20/OECD BEPS project, which is due to report finally on October 5, 2015.  The Treasury Department has consistently debated the proposed use of a ‘special measure’ resembling the commensurate-with-income standard to effect post-transaction value adjustments in the case of “hard-to-value” intangible assets.  Temporary 482-1 Regulations issued September 14 however attempt to reconcile intangible asset expatriation transactions under Section 482 and Section 367.

Let’s hope that Altera has served as a useful caution to the OECD BEPS project, and shown that that country rules that depart from the arm’s length principle and that are based on the OECD transfer pricing guidelines will be subjected to litigation challenge, with the possible unwelcome result that the general OECD guidance will be undermined. 

If the arm’s length principle is to survive, as is widely expected, careful analysis and empirical support will be critical to supporting a transfer pricing position.  The beliefs of taxpayers and tax administrations about what transacting parties would or would not do, mechanical or categorical rules, and simple statements of principle may well lose out to good evidence more often in the future.

 *Portions of this article appeared in a co-authored article published in the Ruchelman PLLC August 2015 Insights newsletter.

Rowing with only one ‘or’ in the water, as the interpretation of section 247(2) remains unclear

On June 12, the Federal Court of Appeal issued its decision on a motion brought by Cameco in an ongoing transfer pricing case.

The Federal Court of Appeal ordered the Crown to communicate to Cameco “its position as to the arm’s length price or prices at which Cameco and CEL ought to have purchased/sold uranium transacted between them during the 2003 taxation year or a formula which allows Cameco to identify this price or these prices”.  One would normally assume that in a transfer pricing dispute, the positions of CRA and the taxpayer on the price of the transaction in question would include an indication of the price each side believes is arm’s length.  The Cameco case shows that in the topsy-turvy world of Canadian transfer pricing, this cannot be assumed.

The CRA reassessed Cameco under paragraph 247(2)(a), paragraph 247(2)(b), Section 56(2), and by alleging the Cameco transaction structure was a “sham intended to conceal the fact that all income earning activities were performed in Canada”.

Section 247(2) allows the CRA to make an income or capital adjustment where paragraph (a) or paragraph (b) applies.  Paragraph (a) applies where the terms and conditions of the transaction or series are non-arm’s length.  Paragraph (b) applies where the transaction or series meets two conditions, namely: (i) the transaction or series would not have been entered into between persons dealing at arm’s length and (ii) the transaction or series “can reasonably be considered not to have been entered into primarily for bona fide purposes other than to obtain a tax benefit”

Where only paragraph (a) applies, the terms and conditions made or imposed, paragraph (c) then applies to re-price the transaction or series.  Where paragraph (b) applies, paragraph (d) then applies to replace the original transaction or series with a different transaction or series that would have been carried on between arm’s length parties, and then use the price from that different transaction or series to price the related-party transaction.

We could just describe the relationship between paragraphs (a) through (d) as follows:

If (a) implies (c) and (a) is true, then (c) is true

If (b) implies (d) and (b) is true, then (d) is true

If (b) implies (d) and both (a) and (b) are true, then (d) is true

But if a company or partnership were to follow the logic above, it may be out of step with the Tax Court of Canada in General Electric Canada and current CRA assessing practice as evidenced in a July 6 proposed reassessment of 2005 to 2010 taxation years of Silver Wheaton.  These interpretations of Section 247(2) suggest that both (a) and (b) can be true in a certain circumstance or that the distinction between them is blurred, opening what appears to be a route to replace a mispriced transaction with a different transaction as opposed to changing the price of the original transaction as follows:

If (b) implies (d), and either both (a) and (b) are true or (a) is the same as (b), then (a) implies (d)

Meeting the two-part test of paragraph 247(2)(b) is difficult, as is making the argument for an arm’s length price under 247(2)(c).  A logical bypass of paragraphs 247(2)(b) and (c) (or at least 247(2)(b)) is therefore a solution that CRA may find worthwhile to attempt.  This may be an attractive option when compared with a legislative amendment to include a rule with two ‘ors’ – (a) or (b) or ((a) and (b)) – that sets out the consequences to taxpayers of meeting these conditions.

Cameco is defending its position vigorously through uncertain territory.  In fairness, the CRA and the Tax Court are not alone in experiencing difficulty reasoning through the distinction between substituting an arm’s length price and substituting an arm’s length transaction when non-arm’s length conditions are found to have been used to determine taxable income.  This issue is the subject of the hotly debated and incomplete BEPS Actions 8, 9 and 10.

As we wait for the Cameco trial and decision, companies should treat Section 247(2) transfer pricing proposals of zero adjusted price, zero adjusted profit, and claims about what arm’s length parties would or would not do with professional skepticism, and consult a transfer pricing advisor.

A proposed treatment for HTVI

If you or your clients suffer from HTVI, or ‘hard to value intangibles’, news of a promising treatment has been announced by the OECD Centre for Tax Policy and Administration.[1] 

Back in the heady days of the emergence of Web 2.0, many young companies and partnerships inadvertently contracted HTVI in transactions involving the sale or cost sharing of new technologies and other intangible assets whose future revenues and cash flows were necessarily difficult to forecast.  More recently, HTVI has been singled out as being one of the leading causes of BEPS, the current affliction of the international tax system.

Many previously untreatable cases of HTVI have been hotly debated in transfer pricing audits, between Competent Authorities, and in the Courts, and involve asset valuations and assumptions that (to some, at least) now resemble home remedies.  Since Web 2.0 has established itself in the business world, hindsight has become very clear to many tax authorities.  As a practical matter, tax authorities often do not have the same information as a company owing simply to the lack of experience with the business, and hindsight is used to diminish this information asymmetry during audits or exams to some extent.

The discussion draft recommends that multinational companies look for evidence in independent transactions [for example in agreements that may have been entered into previously by the Company, or in agreements between similarly situated independent parties that are available in the public domain] of the treatment of uncertain future events.  In the event that independent parties to an agreement view the future as highly uncertain, related parties are encouraged to follow these examples and adopt similar terms used to manage the effects of uncertainty, for example: agreements with a shorter term or that include price adjustment clauses, milestone or contingency payments, or stepped royalty rate schedules.  Some evidence from arm’s-length agreements of events that trigger contract renegotiation when it is clear that the initial agreement was entered into at a time characterized by uncertainty is also suggested as a way to inform the structure of a related party agreement, the terms of that agreement, and the future obligations of the contracting parties to revisit terms.

Evidence of the form of independent agreements, as well as the ability to discern a foreseeable event from a truly unforeseeable event are two things that become critically important in view of the discussion draft.  It follows that the difference between expected profit from the exploitation of an intangible asset and actual profit can differ at arm’s length, provided there have been demonstrably unforeseeable events in the period following the intangible asset transaction.

To become immunized against the future adverse effects of HTVI, the discussion draft proposes that taxpayers provide a full explanation of all forecast inputs and assumptions used at the time of the intangible asset transfer, and explain how the risks expected to be incurred by the contracting parties are incorporated into the forecast.  This explanation should include a comprehensive “consideration of reasonably foreseeable events”.  The second and final step recommended is to document differences between forecasted and actual outcomes in subsequent years, and to show that these differences are the result of developments unforeseeable at the time of the intangible asset transaction.

Though more burdensome, these recommendations do reflect a good measure of common sense, especially if some reliable evidence of contractual terms agreed between independent parties can be found.  Of some concern however are the examples of a natural disaster and a bankruptcy as unforeseeable events in the discussion draft.  The list of surprises or unforeseeable events in business is considered by many to be longer than the various types of acts of G-d or business failures.  One only need look at the long lists of risk factors listed in SEC filings to appreciate the diversity of events that a business might consider unforeseeable.

A recurring theme of the BEPS Project as it relates to transfer pricing matters is the availability of relevant and reliable data upon which to base comparability analysis and adjustments.  The HTVI discussion draft reprises this theme, as the proposed procedures and tests depend critically on the presence and quality of contractual and other information.  As is noted in the discussion draft, this type of information is difficult to find.  Though information exists, it must be recognized that much of the building of the digital economy has been funded privately, making the terms of agreements that are relevant to modern businesses significantly harder to come by.  While this constraint operates on both companies and tax authorities, the approach proposed by the discussion draft places the initial burden of proof on the taxpayer.  Availability of examples of intangible asset transactions and contractual terms at the time of the transaction, as well as examples found after the conclusion of the transaction will become increasingly important to supporting valuation assumptions and documenting the entirety of the valuation process.  Unresolved by the discussion draft is the relevance of the timing of the transfer pricing documentation due date compared with the transaction date when determining which data were available to the taxpayer under relevant country law.

Clearly the discussion draft views opportunistic asset valuations as abusive.  In view of this position, the consistency of the forecasting methods and assumptions employed by companies and their advisors across time and between transactions will become the subject of transfer pricing scrutiny.  A company may be at risk of a transfer pricing adjustment if, for example, a tax authority obtains records of both (1) a valuation calculation prepared in respect of a Year 1 transaction of Intangible Asset Type A that assumes the intangible asset’s economic life is X years in duration, and (2) a buy-in payment valuation in respect of a Year 3 transaction of Intangible Asset Type A that assumes the intangible asset’s economic life is Y years in duration.  It will be ever more important to document the analytical process that is followed to determine the accuracy of assumptions[2] and the standard employed to identify and model uncertain events.

For some, the promise of a cure[3] will present an opportunity to vaccinate against certain strains of future transfer pricing uncertainty (once clinical trials are complete and legislative approvals have been granted, of course).  Unfortunately for others, a tax authority may well administer the treatment involuntarily, at great expense, and without regard for possible side effects.

[1] Discussion draft on arm’s length pricing of intangibles when valuation is highly uncertain at the time of the transaction and special considerations for hard-to-value intangibles, [OECD, June 4 2015]

[2] Assumptions that consist of statements that reference the analyst’s or valuator’s experience would no arguably longer be acceptable.

[3] The OECD Committee for Fiscal Affairs has not agreed a consensus definition of the cure for HTVI

Taxing the returns to intangible assets – will tax authorities know ‘value creation’ when they see it?

One hallmark of good guidance is clarity of terms and their meaning.  The OECD BEPS Project has adopted the term ‘value creation’ to represent, insofar as we can determine, the outcome of functions, risks, assets, funding and the development, enhancement, maintenance, protection and exploitation of valuable intangible assets.  The aim of several of the BEPS Actions is to create tax policy guidance that will enable multinational companies and tax authorities to better connect the tax jurisdiction in which ‘value creation’ occurs and the taxable income attributed to that jurisdiction.  Despite hundreds of pages of published draft guidance, which frequently employ the term ‘value creation’, the term remains undefined.  Troublingly, commentators in the tax community and tax authorities are beginning to use the term in a manner that implies its meaning is understood.

A clear definition is urgently needed, given double taxation often results from fundamental differences in opinion between companies and tax authorities.  If, as many expect, one of the outcomes of the BEPS Action Plan will be more litigation and double taxation cases, and a greater case burden for the already-strained Mutual Agreement Procedure and Advance Pricing Agreement personnel at many of the world’s tax authorities, an undefined or ill-defined term central to the workings of the international tax system will be sure to make difficult conditions worse.

The various BEPS Project drafts suggest that value creation is alternatively something like consumer surplus, the outcome of ‘key value drivers’, an anticipated outcome as opposed to an actual one, and the result of the performance of functions.

In business strategy and economic terms, value creation (or value added) is assessed only by an end customer or consumer.  Value is thought of as the difference between the cost of production and the benefit to a consumer derived from the product, service or right.  Value in this context is not profit.  Companies use labour, capital and intangible assets in production processes to produce or create value for customers, but if customers do not buy or do not exhaust all consumer surplus, all value is not captured by the firm in the form of profit.  More value can be captured than is created – just think of non-competitive pricing.  It is therefore one thing to correlate value creation with firm profit, and entirely another to correlate value capture (producer surplus) and firm profit. 

The OECD’s intangible assets draft also suggests that the factors contributing to value creation are easily separated and their marginal contributions or average contribution shares are measurable.  While this allows a new justification for the use of profit split methods found in other BEPS Project drafts (and a drift away from the arm’s length principle), the reality is that factors affecting value creation are most usually enumerated, and not individually measured or measurable in a robust way.  As value creation is measureable at one point – at the end of the value chain – the mystery of the quantification of marginal additions to value along the chain, always central to the transfer pricing question, remains unsolved.

Terms like ‘value creation’ are common parlance in the business world, but less so in the tax world.  It is not a foregone conclusion that all businesses think of value creation or value added in the same way.  It’s also not clear that tax authorities will not rush to substitute ‘profit’ for ‘value creation’ and demand income tax on an unanticipated share of global profit, somehow measured.  With the current policy focus on taxation at source, often the jurisdiction of the end consumer, ‘value creation’ seems to be a telling choice of term. 

As long as the definition used by businesses and tax authorities is the same, then we can be assured a comprehensible series of policy changes.  Without clear meaning, companies can’t be so sure.

Yes, Virginia, there is a Santa Claus – BEPS Action 11 is still searching for quantification of corporate income tax lost to BEPS

Of all the actions in the G20 and OECD Base Erosion and Profit Shifting (BEPS) Project Action Plan, Action 11 always seemed somewhat out of sequence.  Action 11 is preceded by ten other action items that propose changes to the international corporate tax system (many of these changes have to do with how to calculate a document a transfer price) that will lead to the payment by companies of their “fair share” of corporate tax.  Action 11 proposes to measure how far the current system leaves the world’s governments short of their “fair share” owing not only to incorrect transfer pricing, but also to other forms of base erosion and profit shifting.  Businesses have been asked thus far by the G20 to believe in Santa Claus – to accept that there are large problems with the transfer pricing regime, so that the OECD can get on with the urgent work of fixing things.

On April 15, the OECD released its draft Improving the Analysis of BEPS, making clear that the question of how to measure the loss of tax revenue from BEPS-afflicted transfer pricing activity is still very much unanswered.

Estimates of the loss from BEPS vary widely.  More fundamentally, there is no consensus yet on whether it is the lost tax revenue or the lost economic output that should be measured.  Looked at from a lost tax revenue standpoint, the most reliable estimates with which I am familiar are USD $22 billion (NBER) and USD $70 billion (MCSI).  Though not insignificant in nominal terms, these loss amounts are small when compared with world tax revenue or national income.

It becomes clear however that the political reality of our time will cause economists to wait for tax authorities to gather corporate tax return data to use to estimate the extent of lost tax revenue from BEPS.  The Action 11 draft talks about what should be measured using existing micro-level or firm-level data collected in tax returns.  Tax authorities will get a more thorough data set to work with from multinationals with sales greater than €750 million thanks to the Action 13 country-by-country reporting requirement expected to come into force in 2017 in many countries.  This will influence roughly 1700 US taxpayers, in addition to others. 

Given the ambitious timetable and state of progress of the OECD’s reforms, companies could soon find themselves in a position of having to comply with new transfer pricing legislation without seeing the proof that there is a compelling public finance rationale for this legislation.  Perhaps companies cannot be blamed this time for having been “affected by the skepticism of a skeptical age”, just like the young friends of Virginia O’Hanlon in 1897.

Transfer Pricing Litigation from A to Z

A number of transfer pricing cases, many with potentially significant precedent value and tax provision consequences, are either at trial or proceeding to trial in U.S. courts.  We selected two interesting cases, Altera and Zimmer, to brief and also offer our transfer pricing commentary.[1]

Altera Corp.

Altera Corp.[2] is a California-based manufacturer of programmable semiconductors and related products.  It has sales of $1.8 billion world-wide.  The taxpayer petitioned the U.S. Tax Court, challenging adjustments in the amount of $96.6 million, most of which relate to the inclusion by the I.R.S. of costs associated with employee stock options in its cost-sharing agreement (“C.S.A.”) with its Cayman Islands affiliates for years 2004 through 2007.

The taxpayer’s challenge to the adjustments considers the validity of the 2003 cost-sharing regulations,[3] the Violation of the Administrative Procedures Act, and the legal standard of review.

Altera claims that the 2003 Cost Sharing Regulations, which are amendments to the 1995 Regulations, violate the arm’s length standard by requiring the related parties to share stock-based compensation, a transaction that is not undertaken by unrelated parties.  Therefore, Treas. Reg. §1.482-7(d)(2), which requires the inclusion of stock options in the cost pool, is invalid as a matter of law because it is inconsistent with the arm’s length standard set forth in Treas. Reg. §1.482-1(b)(1).  In fact, the intention of Code §482 is to achieve tax parity, which can only occur if the activities of unrelated parties are considered.

The I.R.S. claims that the sharing of options costs is governed by the commensurate-with-income-standard, which does not require that third-party activities be considered, but only that it “achieves an arm’s length result.”  Accordingly, the behavior of unrelated parties isn’t relevant when determining whether a taxpayer’s cost-sharing transaction terms are consistent with the arm’s length standard, and it can be determined in any way as long as the desired result is achieved.  The I.R.S. position is that stock-based compensation is an economic cost that must be included in the pool, otherwise income would be distorted.

Judge Marvel asked the I.R.S. how they could analyze an issue to determine if it achieved an arm’s length result if it did not take into consideration what uncontrolled parties were doing in the same situation. In response, the I.R.S. said that tax parity is achieved “if you reflect true taxable income.” Therefore, taxpayers that follow the qualified cost-sharing regime achieve parity.  Marvel replied, “It sounds to me like you are saying the only relevant standard is the commensurate-with-income and not the arm’s length standard.”

Marvel further noted that while the preamble to the 2003 Regulations states that evidence submitted by stakeholders was not sufficient, the agency failed to explain why it came to that conclusion. Furthermore, the record does not support the I.R.S.’s position. “Shouldn’t there be something in the rulemaking record that supports your belief that the failure to share stock-based compensation leads to distortion?” Marvel asked.

Under the Administrative Procedures Act (“A.P.A.”), a final rule cannot be enforced unless it is the product of “reasoned decision making” and is “consistent with the underlying statute it is designed to implement.”  Altera pointed out that the I.R.S. in fact adopted the stock-based compensation provision over the objections of multiple stakeholders, who testified that no unrelated party ever shares such costs in development deals.

Altera says that the Section 482 regulations require the I.R.S. to rely on the arm’s length standard, but the administrative record does not show that the sharing of equity-based compensation ever occurs among unrelated parties.  It goes on to contend that the I.R.S. is obligated under the A.P.A. to consider that record.  If not, the regulation is not the product of reasoned decision-making, the regulation cannot pass review.

With regard to the final issue of the legal standard of review, the I.R.S. claims that the 2003 regulations meet the two-step test set out by Chevron[4] and is further supported by Mayo,[5] which held that agency rules deserve deference from reviewing courts because the formulation of the policy requires “more than ordinary knowledge respecting the matters subjected to agency regulations,” so as long as it is reasonable.

Altera claims that the standard should satisfy Chevron and Motor Vehicle Manufacturer’s Ass’n,[6] in which the Supreme Court held that in amending a regulation, an agency must examine “relevant data” and “articulate a satisfactory explanation.”  If an agency fails to do so, the change is arbitrary and capricious and cannot be upheld.  Altera therefore claims that the change from the 1995 Regulations to the 2003 Regulations cannot meet this standard and should not be upheld.

The Altera case is one of a number of cost-sharing cases in process.  It is unique however in that it is, for all intents and purposes, a retrial of the issue of stock option expense inclusion in a C.S.A. as decided in Xilinx.[7]  Considerable evidence of the behavior of independent signatories to joint technology development agreements was offered in amicus briefs and motions during the Xilinx trial proceedings.  The evidence submitted by trade groups and experienced industry participants supported the notion that arm’s length parties do not share stock option costs.

The I.R.S. position will require clarification of the meaning of the arm’s length standard.  Does the standard apply to make parties transact as arm’s length parties would, or cause transacting parties (or one transacting party, usually known as the tested party) to report an arm’s length outcome (in this case income)?

Given that an apparent shortcoming in the I.R.S. position is its failure to adequately consider the actions of uncontrolled parties in the same situation, we considered the general definition of costs in one of the most R&D intensive industries – defense.

Defense contractors are required to account for their costs in conformity with the Federal Acquisition Regulation (“F.A.R.”) Cost Principles.  One of the elements of cost that is allowable under a cost plus R&D contract is stock option expense incurred as a result of the options issuance to employees carrying on the specified R&D activity.  Stock appreciation rights are treated like options for F.A.R. cost purposes under these rules.  Stock options only have a positive cost attribute if the option is in the money on the issue date (i.e., the first date on which the number of units and the option price are known), implying that stock option cost is not always positive (if in fact the F.A.R. Cost Principles are appropriate guidance under Code §482).

Determining whether the F.A.R. Cost Principles are relevant to the pricing of joint development, joint venture, or cost-sharing agreements requires an analysis of comparability of attributes of the agreements carried out under Treas. Reg. §1.482-1(d).  It is this comparability standard that the I.R.S. contends is of relative unimportance when contrasted with the commensurate-with-income standard in the case of Altera.  Whether either an accepted standard or evidence of the behavior of third parties, such as the F.A.R. Cost Principles, is persuasive evidence in the view of the courts remains to be seen.  As always, transfer pricing matters are won and lost on some combination of legal analysis and empirical evidence.

Zimmer Holdings Inc.

Zimmer Holdings Inc.[8] is a publicly traded company based in Warsaw, Indiana, with worldwide operations and annual sales of $4.4 billion.  Its Dutch subsidiary, Zimmer Manufacturing B.V., produces medical products through its Puerto Rico operations. Zimmer operated in Puerto Rico during a time when tax benefits were provided to domestic corporations qualifying as Code § 936 corporations. When the tax benefit was repealed, the standard practice was to migrate the business in Puerto Rico to a Dutch subsidiary corporation. In recent years, the I.R.S. has challenged the tax-free treatment claimed for the migration.

Zimmer is challenging income adjustments made by the I.R.S. of $228.5 million related to the licensing of its intangibles to its Dutch subsidiary, claiming that the adjustments made by the I.R.S. are incorrect and no tax is due for the years 2005 through 2007.

The I.R.S. has taken three separate positions.  The first addresses the transfer pricing adjustments under Code §482.  Zimmer claims the adjustment is incorrect because the intercompany pricing is arm’s length.  The intercompany agreements provide that Zimmer Manufacturing B.V. assumes all risks associated with the production of medical products and indemnifies the parent company for all liabilities, losses, claims, and costs.

Though at opposite ends of the docket’s alphabet, the Zimmer case shares at least one important trait with the Altera case from a transfer pricing perspective.  The evaluation of comparability under Code §482 may well become part of the arguments of both parties and be instructive to the decision.

Where we considered the availability of third-party conventions on stock option expense treatment in the circumstance of Altera, the terms in agreements between independent licensors and licensees may become relevant in the Zimmer case.  While a review of the contractual terms pertaining to risk in the Zimmer Manufacturing B.V. agreement against other licensing agreements may lead to the conclusion that the Zimmer dealings occurred at arm’s length, the actual risks incurred by the parties and the economic circumstances of the parties in the context of the intercompany licensing transaction may in fact have departed from the intent expressed in the agreement.  In this case, substance determines the treatment of the transaction for transfer pricing purposes.  Also relevant may be evidence from arm’s length contracts and other evidence of the outcomes of commercial arrangements, which may or may not accord with the actual conduct of the parties.

Intercompany agreements are essential to have in place in the case of intangible assets transactions.  Agreements evidence the intent of the parties, and are often the first line of defense in a comparability dispute.  We expect the Zimmer case may, in some part, be decided on the basis of comparing intent as expressed in the intercompany agreement with actions as properly evidenced.  Cooperation and communications between tax function leaders in companies and their operations and legal colleagues go a long way, in our view, to making sure form matches substance.

The second position regarded alternative adjustments under Code §367(d).  Zimmer claims that that §367(d) does not apply as there were no transfers specified intangibles under Code §936(h)(3)(B), specifically goodwill and workforce-in-place.  Zimmer also claims in promulgating the regulations under Code §367(d), the I.R.S. violated the Administrative Procedures Act.  Additionally, it maintains that the Code §367 allegations are “internally inconsistent” because they apply royalty rates to an erroneous revenue base.

As a third alternative argument, the I.R.S. argued that intellectual property was transferred under Code §367(a) and imputes a transfer of $1 billion in underlying intangibles from the U.S. parent to the Dutch subsidiary.  The intellectual property license agreements worth $880 million, workforce-in-place valued at $2.5 million, and goodwill valued at $11.6 million, has zero basis and therefore the transfer results in a taxable gain of $998.6 million.

Zimmer argues that no license agreements were transferred to Zimmer manufacturing, so the adjustment based on valuation of property isn’t subject to §367(a)(1).  Further, neither goodwill nor workforce-in-place is subject to §367(a)(1).

Zimmer will be a bellweather case as the I.R.S. has aggressively pursuing all companies that moved legal ownership of intangible property used in Puerto Rican operations to Dutch subsidiaries at the time the tax benefits enjoyed by Code Section 936 corporations were terminated.

[1] We thank Cheryl Magat for her contributions of research and legal analysis, and Stanley Ruchelman for his comments and for the idea that inspired the article.  This article also appears in the Ruchelman PLLC publication ‘Insights’.

[2] Altera Corp. v. Com’r., T.C., Nos. 6253-12, 9963-12, argument on cross motions for partial summary judgment, 7/24/14.

[3] Code §482, Treas. Reg. §1.482-7(d)(2).

[4] Chevron USA Inc. v. Natural Resources Defence Council, 467 U.S. 837 (1984).

[5] Mayo Foundation for Medical Education and Research v. United States, 131 S. Ct. 704, 2011 BL 6645 (2011).

[6] Motor Vehicle Manufacturers Ass’n. of the U.S. Inc. v. State Farm Mutual Auto Insur. Cos., 463 U.S. 29 (1983).

[7] 125 T.C. 37 (2005), rev’d, 567 F.3d 482 (9^th Cir. 2009), opinion withdrawn, 592 F.3d 1017 (9^th Cir. 2010), and aff’d, 598 F.3d 1191 (9^th Cir. 2010)

[8] Zimmer Holdings Inc. v. Comm’r, T.C., No. 19703-14, filed 8/13/12.

Holiday shopping, Canadian retail prices and transfer pricing controversy

If you grew up near the Canada-U.S. border, you may remember getting flyers from U.S. grocery and department stores in the mail, offering bargains for the holidays. The internet now allows instant price comparisons and greater choice for Canadian consumers. For a number of years, Canadians have been telling their governments that our retail prices are unfairly high compared with exchange-adjusted U.S. prices.

The federal government is now preparing to give the Competition Bureau new powers to persuade U.S. multinationals with Canadian retail operations to lower prices or achieve parity (somehow defined). Let’s hope Industry Canada talked to the Canada Revenue Agency before getting too far into drafting legislation, as one unintended consequence may be Canadian transfer pricing controversy.

Exchange rates and transport costs are variable, making a U.S. purchase a relatively better or worse deal than a Canadian purchase at different times. Tariffs, distribution and retail operating expenses, and the profit that Canadian retail subsidiaries of U.S. multinationals must report for tax purposes are relatively more fixed.

If a Canadian subsidiary sets its transfer prices for goods purchased from its U.S. parent by reference to its operating margin (as it not uncommon), a lowering of Canadian retail prices by government fiat will (all else being equal) lower the taxable income of the Canadian subsidiary. U.S. parent companies will be faced with a choice between keeping their customers and the Canadian government happy, and keeping the CRA happy with its subsidiary’s transfer prices.

Good company managers, who don’t allow the tax tail to wag the business dog, will likely vote to keep their customers happy and avoid bad Canadian press. A PR victory will come at the expense of either the U.S. treasury (if the parent cuts its transfer prices to maintain Canadian subsidiary profit), or the Canadian treasury (if the parent keeps its transfer prices the same, or does not make the subsidiary whole for its loss)^[i]. When one or both tax authorities lose, there are more audit disputes, Advance Pricing Arrangements to monitor and renegotiate, and double tax cases to work out through our tax treaty with the U.S.

Here’s hoping the Canadian retail market remains worth the transfer pricing effort for many of our foreign-owned retail establishments.

^[i] We’re feeling the spirit this holiday season, and we know you like numbers. Here are some numbers to illustrate the argument:

Scenario A is the current state. Q*C(TP) is Canadian subsidiary Cost of Goods Sold and depends on the transfer price C(TP) charged by the parent. SG&A includes salaries, and let’s assume CRA has settled a prior-year audit based on a 5% operating margin. The company continues to follow this guidance (correctly, let’s assume) when doing its annual transfer pricing analysis and documentation under paragraph 247(4)(b) of the Act.

When the retail price p falls by 17% in response to government consumer protection regulation, and sales remain at q=100, the company has some choices to make. Under scenario B, the parent decreases C(TP) to maintain gross profit and the subsidiary also decreases SG&A (perhaps by reducing Canadian payroll) from 60 to 50. This maintains the Canadian subsidiary’s net profit ratio. It remains to be seen whether the IRS believes the decline in C(TP), representing a reduction in U.S. company profit, is an arm’s length outcome.

Suppose the U.S. status quo view on the transfer price wins out, and C(TP) remains the same at 225. Sales remain at q=100. Despite cost-cutting in the short-term by the Canadian subsidiary, a loss of 25 results. We expect the CRA may question this extreme result, and argue for an operating profit that is within an arm’s length range after adjustments for comparability. Good written documentation of the facts and circumstances will help explain the position taken.

Free In The Harbour? Draft guidance on transfer pricing for low value-adding services released by the OECD

As part of its work on Action 10 of the OECD/G20 Base Erosion and Profit Shifting (BEPS) Action Plan, the OECD released a draft revision to Chapter VII of its Transfer Pricing Guidelines on November 3.

The draft revision broadly aims to limit deductions for management fees and head office expenses in receiving jurisdictions. Management and headquarters services are viewed by the OECD to be problematic from a pricing information standpoint, given such transactions are viewed to occur rarely between independent parties. Using an approach reminiscent of the IRS services regulations re-write in 2009 and implementation through IRB 2007-3, the draft introduces a new category of services termed “low value adding”.

Meeting the “low value adding” definition is accomplished by the service type either being on one list or not being on another. Other measures proposed, such as determination of cost pools, demonstrating benefit, specific services, drafting service contracts, and allocation methods or “keys”, have been good transfer pricing practice worldwide among practitioners for many years and are simply written down for all to see.

The significant development in the draft is the proposal of a safe harbour profit mark-up of strictly between 2% and 5% on low value-adding services costs. This guidance does alleviate the need to research a services mark-up, and the trade-off with other proposed compliance steps seems on balance to benefit taxpayers in the long-run. How to pick a point in the range of 2% to 5% is left unexplained.

Some structural inconsistencies with other jurisdiction rules will need to be managed, however. The U.S. 482-9 Regs that use a 7% low mark-up threshold instead of 5%, and service centers or service hubs in low-cost jurisdictions (i.e. India, Poland, and the Phillipines) that charge arm’s length mark-ups greater that 5% by reference to local market conditions are examples of areas of conflict to watch.

Like the whales swimming free in modern harbours in the Stan Rogers tune, will simple or ‘low value adding’ services be accepted by the CRA if priced under the draft Chapter VII and spared the traditional harpoon of Section 247(2)(a) of the Act?

The CRA has been clear that transfer pricing safe harbours are not crossing our borders any time soon, and that a ‘facts and circumstances, case-by-case’ approach is the way to go. As Canada is a net importer of service fees, it will be interesting to hear CRA’s views on the Chapter VII draft. A profit safe harbour would simplify compliance and lead to less disputes over the reasonableness of a mark-up on basic service fees, and this may well provide an efficiency benefit to tax authorities like the CRA. Efficiency gains may however be limited in practice given service fee disputes often center around the benefit received from the service, and what services were actually rendered. CRA’s position has traditionally been that a mark-up should not apply on support or management services.

In case you were wondering, “Giving advice on tax matters” is listed as an example of a low value adding service. Comments are due from interested parties and their tax advisors in January 2015.

Should a company have a transfer pricing agreement?

‘Transfer pricing agreement’ is sometimes the way a company owner or manager will describe the document that is needed to prove a bona fide transaction and arm’s length transaction terms to a tax authority. The term ‘agreement’ is often easily substituted for the term ‘documentation’ in conversation. The term ‘documentation’ means something different to tax practitioners in a transfer pricing context, and is synonymous with the requirements of paragraphs 247(4)(a) and (b) of the Act, Part 7 of IC 87-2R and TPM 09 posted on the CRA website (not to mention other countries’ transfer pricing documentation requirements).

Recent developments suggest that an inter-company agreement and documentation are both important parts of explaining and supporting a transfer pricing tax position on audit.

Perhaps the best reason to have an agreement in place is that the CRA and other tax authorities routinely ask for agreements in audit queries. While tax authorities will often state that related-party agreements lack credibility due to the relationship of control between the contracting parties, an dated and signed agreement often has a the effect of reassuring an auditor of the timely exercise of forethought, provides a clear expression of intent on the part of the company, and is generally suggestive of good records and record-keeping procedures.

Several recent transfer pricing decisions from the Tax Court of Canada have relied on the analysis of the terms of intercompany agreements against arm’s length comparables.

Written agreements tend to demonstrate adherence of transaction terms to commercial norms, if only in their legal form. Even though a related-party transaction often has no direct market analogue, being able to describe the terms of a related-party transaction in the familiar form of an agreement supports the notion that the related-party transaction is not patently unreasonable.

An inter-company agreement sets out the form of a transaction and the obligations of the parties. Support for a future argument against transaction re-characterization by a tax authority can often be found in an intercompany agreement. As a starting point, the agreement sets out the inputs for a transactional comparability analysis. A well drafted agreement contains clauses that mirror the list of comparability factors found in most country tax law and administrative guidance.

As is the case with any agreement governing a complex transaction, an inter-company agreement should be drafted or reviewed by a lawyer. Inter-company agreements, while neither substitutes for the detailed information contained in transfer pricing documentation nor required by law in many instances, are another tool that companies should use to manage the transfer pricing aspects of international related-party transactions.

‘California Dreamin’ for only $32,500 per year* (*Part I and Part VIII tax, interest, and transfer pricing penalties not included)

Until June 10, companies could only make an educated guess when asked whether their Canadian transfer pricing documentation would protect against a penalty under paragraph 247(3) of the Income Tax Act. The Tax Court of Canada’s decision against the taxpayer in Marzen Artistic Aluminium Ltd. v The Queen (2014 TCC 194) has diminished this uncertainty.

Marzen was a B.C. window manufacturer that entered the California market, benefiting from the building boom of 2000 and 2001. A change in U.S. marketing strategy sent the company (on the advice of its tax lawyers) to a Canadian lawyer and businessman based in Barbados for further advice. The Tax Court determined the price of this advice (even though it was agreed to have been “game-changing” for the business, and included an annual administration fee for a Barbados subsidiary) was an arm’s length price – $32,500, and used it to uphold CRA’s $7.1M section 247(2)(a) transfer pricing adjustment.

Separate and apart from obtaining the advice, Marzen incorporated a Barbados subsidiary (SII) to perform marketing services in respect of the U.S. market. In exchange for these services, the Barbados subsidiary agreed to receive a fee equal to 25% of U.S. sales. A further 10% bonus was agreed for a time as an incentive to close sales. Marketing functions in the U.S. market were performed by SWI, a Canadian company related to Marzen.

Marzen was selected for an international tax audit in 2003, and responded to CRA’s request for documentation within the three months allowed under paragraph 247(4)(a) the Act. The company submitted the following:

• a cover letter from counsel;
• the four inter-company agreements creating the Barbados Structure
• correspondence between SWI and SII regarding increasing SWI’s fees
• correspondence between SII and Marzen regarding the one-time 10% Bonus Agreement;
• a business study of the U.S. market prepared by a Marzen executive

The Tax Court found that the requirements of subparagraphs 247(4)(a)(v) and (vi) were not met. In short, there were no data, methods, or analyses provided to support the transfer prices used in the related-party transaction and no explanation of the assumptions, strategies or policies that influenced the determination of transaction prices. The Tax Court relied on CRA’s expert rebuttal, written by one of its own economists. A $500,000 penalty was upheld by the Tax Court, even though the income adjustment exceeded the $5M threshold by only $25,000.

Marzen is the type of transfer pricing case that CRA has been waiting to take to trial and win for a number of years. We can expect many others in queue to either settle in favour of CRA on the basis of the Marzen decision or build on the 247(3) jurisprudence if they proceed to trial.

The tax motive for the Barbados structure was noted, but had no influence on the Court’s decision. Extensive questions of substance in the Barbados company were raised at trial, most were decided in favour of CRA, but none was directly crucial to the decision.

This taxpayer, which could have become the first Canadian poster child for the G20/OECD BEPS Project under different circumstances, really lost because it forgot to do the math.

A transfer pricing lesson from McKesson?

In late 2013, the Tax Court of Canada released its decision on a transfer pricing matter in the case of McKesson Canada Corporation vs. The Queen (2013 TCC 404). Under appeal was the assessment of Part I and Part XIII tax resulting from a Canada Revenue Agency (CRA) paragraph 247(2)(c) adjustment of the discount rate applied to the face value of several portfolios of McKesson Canada trade accounts receivable sold during its 2003 taxation year to a related corporation resident in Luxembourg. As a result of the Court’s decision, the discount rate that had been reduced from 2.206% by the CRA at the audit stage to 1.013% was found to have an arm’s length value of between 0.959% and 1.17%. McKesson has filed an appeal of the Tax Court decision with the Federal Court of Appeal.

While the greatest share of the Tax Court decision deals with the quantification of various components of the discount rate and the timing of assessment of withholding tax under the Canada-Luxembourg treaty, several more general but useful reminders of good practice can be taken from the decision. These lessons are the focus of this article, and are worth recalling especially at this time of year, when many companies are preparing transfer pricing documentation in advance of the popular June 30 corporate tax return filing due date.

1. State and explain assumptions in documentation, support analytical steps with data and other evidence: The Tax Court was left to examine the facts of the case and conduct its own pricing analysis owing to a number of unexplained assumptions that were made in the McKesson pricing analysis. Averaging periods, risk of default, and discount rates were stated but not explained, and adjustments were made using rough ‘buffer’ factors and measurements of variability.

It is reasonable to expect that CRA and a company can disagree on a transfer pricing method selection or pricing calculation, but it is also reasonable to expect that the reasons for disagreement will become discernible. Without stating assumptions used in calculations (a requirement under paragraph 247(4)(a)(vi) of the Act), a company’s position becomes difficult to explain to the CRA on audit without providing significantly more information and analysis. What’s more, a lack of clarity on assumptions makes it difficult for companies and the CRA to whittle down their respective lists of points in dispute, ‘agree to disagree’ on any specific remaining matters in a fully informed manner, and obtain resolution from Appeals Branch, Competent Authority or the Tax Court.

2. Adequacy of documentation – contribution to understanding of facts and circumstances of the transaction: Not only did the Tax Court spend considerable time conducting its pricing analysis, it was left with the daunting task of gathering and interpreting evidence of the facts of the transaction in dispute from the “reams and reams of documentation entered into evidence”. A footnote hints at the diminishing “will to live” experienced by the author when reading his own reasons for judgment in such a matter.

The ‘reasonable efforts’ test of section 247(3) pertains only to the determination of whether a company is liable for a transfer pricing penalty. A company must decide how much effort to make (and expense to incur) in producing transfer pricing documentation.  At the same time, a company must consider the expected risk and cost of a bad audit outcome and do a sufficiently good job of telling the story of the transaction in its documentation. The business and commercial circumstances in which the transaction was undertaken, the options available to the transacting parties at the time of the commencement of the transaction, and the expected benefit to the transaction participants are all elements of the story of the transaction.  These elements help to support both the existence and the terms of the transaction, both from a business standpoint and in comparison with terms and conditions between unrelated parties.

3. Transfer pricing methodologies – other methods and double checking: The Tax Court decision used a transfer pricing method that can only be described as “Other” when compared with the options presented in IC 87-2R. “Other” methods are not new in Canada. General Electric Capital Canada Inc. was decided with reference to an “Other” method. The facts of McKesson do not necessarily lend themselves to a double check of the “Other” transfer pricing method, but the profit consequences of the transaction in dispute were quite clear. McKesson’s profit turned to a loss as a result of the receivable sale transaction. One lesson from McKesson may be that a double check (if practical) of the primary transfer pricing method with a secondary method (or other test) is advisable.
4. Timing of analysis (pre- or post- transaction): Though this factor was not essential to the CRA’s case, the decision discusses the timing of the determination of the arm’s length discount in relation to the initial sale of accounts receivable and the initiation of the series of monthly A/R sales transactions. It appears that terms were settled and written into the intercompany agreement before the determination of an arm’s length price could be concluded.

Best practice suggests the right time to price related-party transactions is before a transaction takes place. Second-best practice (which is often the only practical option) requires the determination or adjustment of transfer prices before filing a corporate tax return. This way, intercompany agreements, transfer pricing documentation and taxable income can be consistent.

5. Relevance of an intercompany agreement: The Tax Court relied extensively on the agreement between the related parties when determining the terms of the transaction and in conducting its pricing analysis. Though not mandatory, in the case of McKesson the presence of an intercompany agreement contributed significantly to the Tax Court’s decision to examine the pricing of the transaction rather than attributing different terms and pricing to the related-party transaction (re-characterization under paragraphs 247(2)(b) and (d) of the Act). Though the Tax Court’s decision acknowledged McKesson’s international tax structure and a tax motivation to the transaction, the agreement was used extensively by the Court to determine the terms and conditions of the transaction in dispute.

An intercompany agreement is on balance a good thing to have in place, especially in the case of a complex transaction. On the other hand, an intercompany agreement is not a good thing to have if its terms and conditions are not followed, or if it is inconsistent with the transfer pricing analysis and documentation.

6. Transfer pricing reports as advocacy documents: A charge of “advocacy” was levelled at certain of McKesson’s transfer pricing studies or reports cited in the Tax Court’s decision. The description of transfer pricing studies and reports furnished to McKesson as advocacy documents I believe neglects the role that such documents play in the real world of Canada Revenue Agency transfer pricing audits.

There is certainly an identifiable line between blindly advocating for a client’s transfer pricing position in spite of the facts and arguing objectively and transparently for a well-supported tax position under section 247(1) (compliance with the arm’s length principle with respect to pricing) and 247(3) (compliance with the section 247(4) documentation requirements). It is however well known that Canada is home to some of the world’s most acrimonious transfer pricing audits. Proposed reassessments are often unsupported by law, administrative guidance, facts, or figures. If this is the world we must work in, I think companies can be excused for advocating responsibly for an income tax filing position pertaining to transfer pricing.

Put differently, I think a certain amount of responsible, principled and well-supported advocacy or argument for a particular transfer pricing methodology, transaction characterization, or calculation method can lead only to more clarity in positions and therefore more efficient resolution of audit disputes. Definiteness in transfer pricing positions is a rare bird these days.

7. Selection of an advisor or expert: The McKesson decision is not particularly complimentary of the various advisors and experts retained by McKesson. The question of the relative merits of subject matter expertise versus transfer pricing expertise is raised in several parts of the decision.

The Tax Court had to rely on portions of expert testimony and reports, but ultimately found some use for both the subject matter (securitized lending) and transfer pricing (discount pricing) expert input. This suggests that both types of experts are useful, but perhaps also that each has her/his own function. Despite best efforts by each type of expert, it may remain the case that judicious use of more than one expert or advisor is good practice from the standpoint of establishing a well founded position in facts and circumstances, comparability, and application of a well-selected transfer pricing method.

McKesson’s appeal of the Tax Court’s decision to the Federal Court of Appeal may change how seriously companies should take these lessons.  In the short-run, companies should take note of a number of the lessons from the McKesson case when deciding on an approach to selecting a transfer pricing method, applying and supporting a transfer pricing method, and documenting transfer prices to meet the requirements of Section 247.

IRS vs. OECD — how are tax authorities planning to conduct your next transfer pricing audit?

Over the past 15 months, the IRS and the OECD separately published transfer pricing audit and administrative initiatives that will significantly impact the way controlled transactions among related parties are reviewed. These initiatives are consistent with overall concerns raised in the OECD’s Base Erosion and Profit Shifting (“BEPS”) Report. Each stands independently of BEPS and will likely be unaffected by the ultimate action plans implementing the BEPS goals.

View from the U.S. by Bob Rinninsland (All R’s LLC)

Congress has not passed any significant transfer pricing legislation in recent years and U.S. transfer pricing regulations remain essentially unchanged. As a result, the U.S. “best method” rule remains the norm.

However, technical rules are now viewed through a lens coated with the political fallout from testimony before the Senate Permanent Subcommittee on Investigations of several large multinational corporations (Apple, Microsoft, Hewlett-Packard and Caterpillar). The IRS continues to expand its team of transfer pricing examiners and is refining its information exchange and advance pricing agreement procedures.

The IRS has significantly changed the U.S. transfer pricing landscape in the absence of legislation by exercising its administrative authority. It concluded that it needs to develop transfer pricing cases more thoroughly at an earlier stage in the audit process in order to identify and resolve issues without resorting to the appeals process. IRS audit teams are spending more time on advance preparation. Audit teams now regularly research a company’s business and industry and adopt a “big picture” approach to a case in lieu of a straightforward application of transfer pricing regulations.

The drive to develop transfer pricing cases more thoroughly at an earlier stage is viewed by some advisers as an attempt to bypass the importance of the appeals process within the IRS by an emphasis on building a litigation file. This approach significantly changes the dynamics of the audit process.

In February, the IRS released its Transfer Pricing Audit Roadmap (the “Roadmap”) which is intended to provide audit techniques and tools to plan, execute and resolve transfer pricing examinations. The Roadmap anticipates up to a 30 month timeline (6 months of planning and 24 months of audit) for the planning, execution and resolution of a Quality Examination Process or QEP. The QEP is based on certain fundamental assumptions. First, up-front planning is essential. Second, transfer pricing cases are usually won or lost on the facts. Third, a reasonable result under the facts and circumstances of any case should be attained. Finally, effective presentation can “make or break” a case.

The QEP emphasizes early identification and prioritization of transfer pricing issues. This will determine proper staffing and scope of the audit given the anticipated complexity of the case. Regarding the existence of facts to justify an adjustment, the QEP notes that the key in is to put together a compelling story of what drives the taxpayer’s financial success, based on a thorough analysis of functions, assets, and risks, and an accurate understanding of the relevant financial information. The QEP will be looking for scenarios that it believes are too good to be true. Similarly, the QEP notes that the transfer pricing team should avoid adjustments where the taxpayer’s financial results are reasonable and the taxpayer’s transfer pricing method fits its profile.

The transfer pricing team’s working hypothesis will serve as a guide to further detailed examination, subject to the collection of new data. The QEP discourages fishing expeditions and encourages a commitment by the transfer pricing team to address in full the taxpayer’s analysis. In this way, the QEP acknowledges that the taxpayer may have the more compelling position on the issue.

As to effective presentation, the QEP focuses on the notice of proposed adjustment. The QEP intends that the notice should serve as a persuasive argument for the accuracy of the transfer pricing team’s position over the taxpayer’s position. It should contain all of the relevant facts, both good and bad, and should lead to a conclusion that is self-evident. The QEP assumes that a well presented notice of proposed adjustment will increase the odds of early resolution or a favorable result on appeal. Some advisers believe that, this assumption of the QEP appears to be that the transfer pricing team should prepare a position paper that is at least as good as the transfer pricing report of the taxpayer.

While the QEP process may seem reasonable on its face, further consideration raises two key questions:

1. Is this an audit or preparation for litigation? Notable in the QEP detail is an emphasis on documentation of the audit steps taken, facts discovered, preliminary risk assessment, ongoing factual analysis and ongoing coordination with various TPO personnel and counsel. Preparation of a mid-cycle risk assessment to update the initial risk assessment and analysis is considered an important component of the QEP. Finally, participation of the audit team in the appeals process itself with a view towards understanding of the appeals rationale and consideration of future years’ risk assessments could be considered an expansion of normal audit team participation at that level.
2. Is the QEP approach consistent with current transfer pricing law and regulations? Remember that current transfer pricing law and regulations remain the same. The QEP “big picture” view may or may not align with existing transfer pricing regulations that do not necessarily require a focus on overall economic outcomes or financial results.

Nevertheless, the fact that the TPO organization now is the key IRS transfer pricing administrative function and that the QEP represents the TPO’s key transfer pricing enforcement mechanism imply that taxpayers will need to consider the goals and objectives of QEP in managing audits and in establishing or revising their future transfer pricing policies. This is especially true with respect to intangible property. The TPO Director has repeatedly indicated that transfer pricing for intangibles will be the top priority for TPO activities and that the exam approach should consider the overall economic outcomes achieved by the intercompany transactions involving intangibles and not just whether those transactions have complied with specified methods in the regulations. According to the TPO Director, many related party intangibles transactions achieve unrealistic results and would never be observed between independent entities. Whether this view will ultimately prevail may well depend on the quality of the QEP presentation rather than the expectation of the TPO Director.

The IRS appears to be coming around to a new, more modern, strategic approach to tax management involving the systematic assessment of transfer pricing risk and the corresponding targeting of resources and efforts accordingly. The approach is modeled after findings from the OECD’s tax assessment and compliance research over the last 15 years, and programs implemented in Australia and the United Kingdom, thus reflecting the more positive international transfer pricing developments.

This new approach envisions a more engaged, more cooperative style of examination and a greater use of prescriptive tools, including the development of profiles, or templates, of required information and/or outcomes (based on statistical and other metrics), against which taxpayers can be measured and evaluated with prescribed remedial action depending how the company matches up against the profile. Such action ranges from no action, to follow-up questions, and to a more detailed request. The IRS has indicated that they have already developed several profiles.

The expectation, based on experiences in other countries, is that companies that fit the profile in terms of timeliness and completeness will experience a lighter, quicker and less costly IRS examination. On the other hand, the approach is intended to quickly identify issues that can be given greater attention by more resources and more effective resources.

Viewed from a different perspective, the QEP essentially represents a reordering of the decision making process concerning litigation. The effort that will be made in connection with the decision to proceed with a notice of proposed adjustment means that once a decision is made to issue the notice, the role of the appeals officer in resolving transfer pricing controversies will be reduced because the facts gathered by the transfer pricing team will be clear and convincing. The end result is that the risk of litigation assessment by the appeals officer will be perfunctory – a state of affairs well known in Canada.

View from the OECD

Not long before the release of the Roadmap, the OECD released two documents that set out the current guidance to its 34 member states (as well as G20 member states) on pre-audit risk assessment, transfer pricing documentation and country-by-country (“C-b-C”) reporting.

The Draft Handbook on Transfer Pricing Risk Assessment (OECD, April 30, 2013) (the “Draft Handbook”) is a collection of recent country procedures, methods and approaches intended to help tax administrations improve performance. The objective of the Draft Handbook is to promote more efficient audits by tax authorities in order to avoid waste of resources for tax administrators when unsustainable positions result in litigation Competent Authority cases. While there are no mechanical rules prescribed by the Draft Handbook, countries are encouraged to follow regular and structured risk assessment steps. The Draft Handbook is not law, administrative practice, or even necessarily prescriptive in its approach. The OECD makes it very clear that each country will need to develop its own approach to risk assessment.

The intent of risk assessment is to help an OECD member tax authority determine the factual inquiries that it will make during the course of a transfer pricing audit, if a full audit is to be conducted. There is clear reference to the trade-off between the understanding of risk and the extent of information available for review at the risk assessment stage.

The Draft Handbook deals only with recommended pre-audit procedure. Chapter 4 of the OECD Transfer Pricing Guidelines deals specifically with examination practices, albeit briefly.

The January 30 Discussion Draft on Transfer Pricing Documentation and C-b-C Reporting (OECD, January 30, 2014) (“the Discussion Draft”) proposes a working version of a new standard of documentation and country-by-country information reporting that is considerably more extensive than the present Chapter V guidance.

As one of 15 BEPS Action Plan steps taken in a time of fiscal crisis, the Discussion Draft recalls the approach to serious crime in occupied North Africa taken by Captain Renault in the classic film Casablanca: “Realizing the importance of the case, my men are rounding up twice the usual number of suspects.” The volume and utility of the information requested in the Discussion Draft, as well as information security and confidentiality, has been roundly criticized by the tax community. The Discussion Draft states that information submitted to tax authorities (either documentation or the new C-b-C factual and financial reporting) can be used in either the pre-audit or case selection phase of a transfer pricing audit, or can be used in the early stages of an audit for the purpose of focusing such audits on the most important issues. Irrespective of how or if the information will be used, the Discussion Draft calls for more C-b-C reporting information that can be obtained by a tax authority before review of the transfer pricing documentation.

The U.S. developments with QEP have been independent of the C-b-C dialogue and, in fact, U.S. officials have expressed some reservation as to the logic of certain aspects of the C-b-C reporting requirements. This is easy for the U.S. to say, as the U.S. already has in place a robust reporting regime for international business operations of U.S. taxpayers. This regime is an integral part of QEP planning phase which contemplates a detailed tax return review including (i) Forms 5471 and 5472 regarding information on intercompany transactions, (ii) Form 8833, regarding treaty based return positions, (iii) Form 8858, regarding information on disregarded entities, (iv) Form 8865, regarding U.S. controlled foreign partnerships, (v) Schedule UTP, regarding uncertain tax position disclosures, and (vi) worldwide book to taxable income reconciliation Schedule M-3 of the Form 1120. Examination of the overall data requests required by these forms would reveal that a material amount of the information requested in the C-b-C reporting has been compiled. Note though that these forms demand the greatest amount of information from U.S.-based groups. The question arises whether the same degree of information should be demanded of local subsidiaries.

But also at issue are the usual suspects. Rather than setting out a risk-assessment process framework like the Audit Roadmap, the Draft Handbook places emphasis on company and transaction fact patterns that are likely to increase transfer pricing risk. The usual suspects are (among others): (a) transactions with related parties in low-tax jurisdictions, (b) intra-group services, (c) excessive debt and/or interest expense, and the transfer or use of intangibles to/for related parties.

The Audit Roadmap sets out the audit process as means of organizing fact gathering and formation of a theory of a case (facts seeking a theory) rather than a theory seeking facts. We believe this is generally the correct way conduct a transfer pricing examination. To some extent, the increased information requirements of the proposed OECD C-b-C reporting and the prescriptive issues lists in the Draft Handbook promote a theory seeking facts approach to transfer pricing risk assessment. We expect double tax issues between the IRS and the tax authorities of its treaty partners will require further effort and time to align the fact development and robustness to the theory of the case where the treaty partner has reassessed tax based on a usual suspects approach.

Summary

From the IRS perspective, whether the glass is half full or half empty, there will be an expanded access to the IRS audit team and other administrative personnel. Taxpayers may want to closely examine their tax situations in 2014 both historically to open years and prospectively to future years so that they may measure the anticipated effect of the IRS initiatives described above. A robust transfer pricing report that tells a complete factual story and supports the transfer pricing method selected and applied may ultimately provide a quicker, more cost-effective means to resolve transfer pricing issues.

From the OECD perspective, we anticipate that there may be information shortages in certain OECD member countries, but expect that the matter will be solved with the introduction of more focused foreign reporting forms. In a sense, the OECD’s emphasis on information requirements is understandable. Reliable information is required to assess risk and responsibly, select taxpayers, and further select particular tax positions for a robust examination. As the Draft Handbook remains in draft while other BEPS Action Plan items receive attention from the OECD Centre for Tax Policy and Administration, we hope that the Audit Roadmap and other procedural developments will be finalized with double tax minimization in mind.

Multinational businesses with a taxable presence in both the United States and in other OECD and G20 member states should be mindful of the similarities and differences between OECD guidance and IRS field guidance. Areas of difference are relevant to exam approaches, documentation approaches, and differences in the perspective of tax authorities conducting Simultaneous Examination Program audits and Competent Authority negotiations.

Tax authorities and the politicians to whom they report have determined that it is time for countries to take control of their tax borders. Transfer pricing examinations of a new and ambitious sort are to be expected.