3. An overview of transfer pricing in extractive industries

Michael Keen, and Victor Thuronyi
Published Date:
September 2016
  • ShareShare
Show Summary Details

1 Introduction

This chapter provides an introduction to transfer pricing in extractive industries operating in resource-rich developing countries. Managing natural resource wealth is central to efforts to mobilize revenue to achieve sustainable development and prosperity in developing countries (IMF, 2011, p. 57; UN, 2000, 2002). Natural resources (fossil fuels, metals and ores) rose to 22.7% of total global merchandise trade in 2012 and they are the largest sources of government revenue for many developing countries (Boadway and Keen, 2010; WTO, 2014). Developing countries rely substantially on foreign private investment to explore for, develop and extract these resources, and transfer pricing administration and enforcement are an integral part of maintaining host government revenues from resources. Remarkably little attention has been given in leading international transfer pricing guidelines to issues or examples involving extractive industries. Only recently has attention been paid to transfer pricing involving developing countries (IMF et al, 2011; UN, 2013).

This chapter will focus on transfer pricing challenges facing a developing country that is a host for foreign investment in nonrenewable resources, such as petroleum and hard minerals. Generally, the process of exploring for and finding a deposit, developing the well or mine, removing and shipping the resource and closing the depleted site takes long time periods and involves substantial investment early in the project (Boadway and Keen, 2010). Unlike intangibles, the resources are immobile and, because there is limited supply and in some cases the resource has unique attributes, economic “rents” often can be earned from their exploitation (Gruber, 2011, pp. G-9; WTO, 2010, pp. 65–68). Much of the development of natural resources is by private investment under a long-term contractual agreement with the host government that controls the mineral resources.

The private participants in the extractive industries include the world’s largest and most sophisticated multinational companies. From a transfer pricing perspective, the extractive industries present a range of challenges; some are distinctly different from and others are quite similar to the challenges presented by intangible-rich industries that are the focus of current attention by the G20, OECD and United Nations. The final OECD BEPS report on transfer pricing adds five paragraphs to the OECD Transfer Pricing Guidelines on commodities but adds little new (OECD, 2015c, pp. 51–54).

The purpose of this chapter is to provide an overview of transfer pricing issues in a developing country extractive industry context. It identifies difficulties host country governments face in administering transfer pricing rules and considers strategies governments may adopt to constrain tax avoidance within the scope of current international tax law and practice. The primary focus of this chapter is on the legal framework and business context in which transfer pricing issues arise for host countries. This chapter does not address local issues of governance and tax administration.

There is a great deal a host country government can do in the area of transfer pricing to protect its tax base while fostering a climate for investment. The key ingredients for effective transfer pricing enforcement are political will, an effective legal regime, disciplined administration and tough but fair enforcement.

Section 2 describes what “transfer pricing” refers to in the context of taxation and why it is important. Section 3 discusses taxpayer objectives in structuring transfer pricing and host country objectives in administering and enforcing transfer pricing. Section 4 explains why the arm’s length principle is used as the international standard for transfer pricing in related party transactions. Section 5 briefly describes characteristics of foreign investment in nonrenewable resources and identifies challenges from a host country taxation perspective of applying the arm’s length principle to related party transactions in the extractive industries. Section 6 reviews strategies that governments may employ to combat transfer pricing noncompliance with the arm’s length principle. Section 7 concludes.

2 What is transfer pricing and why is it important?

In the context of taxation, transfer pricing refers to the prices that are charged for transfers of goods, services, property or other items of value in transactions between persons that are under common control. “Common control” is a factual concept and can include persons within a family, corporations or businesses that have a common owner or owners and even unrelated persons when one person can direct the action of another person in the context of the transaction in question. The term used internationally for persons subject to transfer pricing rules is “associated persons,” and transactions among associated persons that are subject to transfer pricing rules are referred to as “controlled transactions” (OECD, 2010, p. 23; UN, 2013, p. 474). In this chapter, the discussion will focus on associated persons that are companies in a multinational group or enterprise (MNE).

Why is transfer pricing an important taxation issue? In most business contexts involving unrelated persons, the pricing of a transaction is a zero-sum game on a pre-tax basis; what one side wins the other side loses. If the seller receives a higher price, the buyer loses by paying more and vice versa. Moreover, in a competitive market, the interplay of supply and demand provides an effective price discovery mechanism that tax systems may reasonably rely on.

Where a transaction is between related persons, or persons that, if not formally related, are under common control, there is little adversity of interest between these persons. If the parties to a transaction view their outcomes in terms of the combined result of both sides, the zero-sum nature of the pre-tax pricing decision is unimportant. If, in addition, associated companies that are parties to a transaction are taxed differently, and a price is charged so that there is a lower overall tax, together the two companies are better off. The related or associated person condition for application of transfer pricing tax rules is a proxy for the absence of adverse interests. Absent adverse interest, it is possible to shift tax burden through transfer pricing and share the benefit of a reduced combined tax.

In the context of transactions occurring solely within one country or, more relevantly, within one tax system, transfer pricing acquires significance where the parties to the transaction will be taxed differently. One company is subject to full income tax and the associated company may be operating under a tax holiday. Transfer pricing also can be important where companies are subject to exactly the same taxing rules, but one has a “tax attribute” that will result in a different tax on the transaction. A common example is when one company has a net operating loss for tax purposes that it may not be able to deduct because it does not have enough income but the other company is profitable. In each of these cases, in the absence of other considerations, there is an incentive to lower the companies’ combined taxes by arranging the transfer price so that the companies together pay a lower tax.

One reaction may be despair that transfer pricing concerns must be everywhere since there often are tax disparities among associated companies, both in a purely domestic context involving one tax system as well as in cross-border contexts involving two or more tax systems. There are (at least) three reasons transfer pricing is less frequently a concern in a purely domestic context. First, tax systems tend to be internally coherent, so situations that offer great opportunities for tax advantage are fewer or are intended to provide an incentive to engage in certain behavior. Second, unintended opportunities for advantage tend to be temporary or episodic in nature and do not pose systemic tax risk to the host country. Third, where unintended systemic risks occur, it generally is possible within a single tax system for the host government to take steps to address them.

While the question whether a transfer price is “correct” arises most often with respect to transfers of value across the borders of two or more countries, the resource sector, particularly in developing countries, is an exception to this general observation. Some royalty and most resource rent taxes apply to the value of the mineral at the point of extraction – the mouth of the mine or the wellhead. To calculate the value at extraction, the income must be calculated at the point of extraction and thereby separated from the value added after that point. Another example arises when a company operates several mines in one country through nonconsolidated corporations and uses centralized services charged to each separate corporation (see Chapter 4, this volume, Calder).

The potential for tax advantage, however, is common in cross-border controlled transactions. In even the simplest two-country case it is likely that there are disparities in tax rates, and there may be tax base differences as well. Where two countries impose tax on the same transaction, the countries always have different revenue interests. Even if there is no adversity in interest among commonly controlled companies involved in the transaction, because the effective tax rate is the same in both countries involved in the transaction, the transfer pricing nonetheless is a zero-sum game as between two taxing countries. If one country has more revenue in its tax base as a result of its taxpayer receiving a higher price, the other country’s tax base is reduced by the higher cost and therefore smaller taxable income of its taxpayer and vice versa (IRS, 2006).

Increasingly transactions and ownership of operations are structured through legal entities that do not bear a meaningful corporate tax because they are located in countries that do not tax income or facilitate very low effective tax rates on the income. These include countries that do not tax income (Bermuda, the British Virgin Islands, the Cayman Islands, the Channel Islands), that purport to tax but do not (Cyprus, Malta), that allow structures that deliberately under-tax (Ireland, Luxembourg, Mauritius, Switzerland) or that do not tax income outside the jurisdiction and either allow holidays from tax or impose low tax on domestic income (Hong Kong, Singapore). In these cases, only one country involved in the transaction may have a genuine tax interest, so there is little or no intergovernmental dispute.2 The benefit from income shifting in those cases goes almost exclusively to the taxpayer. This chapter will focus on contexts in which the taxpayer benefits from income shifting through transfer pricing. The potential stakes may be seen in the following examples.

Example 1

Assume that Company A in Country X owns all of the stock of Company B in Country X and Company C in Country Y. Assume that Country X taxes income at a 30% rate and Country Y taxes income at a 20% rate. Further assume that Company B mines coal at a cost of 100 and sells the coal to Company C in Country C for 160 for a profit before tax (PBT) of 60 (freight cost is ignored). Company C, in turn, incurs 20 of selling and marketing expense and sells the coal to customers in Country C for 200, for a pre-tax profit of 20. Under this structure, Company B would pay 18 of tax to Country X (60 × 30% = 18) and Company C would pay 4 of tax to Country Y (20 × 20% = 4). See Figure 3.1.

Figure 3.1Transfer pricing between two countries – example 1

From the perspective of the Company A controlled group as a whole, the total pre-tax income in Example 1 is 80 (60 earned by Company B and 20 earned by Company C), and total taxes are 22 (18 paid by Company B and 4 paid by Company C). The Company A controlled group’s overall effective tax rate, determined by dividing the group’s total taxes paid (22) by the group’s pretax income (80), is 27.5% (22/80 = 27.5%).

Example 2

Assume the same facts as in Example 1, except that Company B sells coal to Company C for 140 (instead of 160). Company B will have a pre-tax income of 40 (instead of 60) and have a Country X tax of 12 (40 × 30%) instead of 18. If Company C again incurs 20 of marketing costs and sells the coal to Country Y customers for 200, Company C would have pretax income of 40 (instead of 20) and the Country Y tax would be 8 (40 × 20%) instead of 4. See Figure 3.2.

Figure 3.2Transfer pricing between two countries – example 2

The disparity in tax rates between Country X (30%) and Country Y (20%) creates the opportunity and the incentive to shift profit from Country Y to Country X through changing the transfer price.

From the perspective of the Company A controlled group as a whole, the total pre-tax income in Example 2 remains 80 (40 earned by Company B and 40 earned by Company C). However, by lowering the transfer price paid by Company C to Company B from 160 to 140, total taxes are reduced to 20 in Example 2 (12 paid by Company B and 8 paid by Company C) from total taxes of 22 (in Example 1). The Company A controlled group’s overall effective tax rate, determined by dividing the group’s total taxes paid (20) by the group’s pretax income (80), is 25% (20/80 = 25%) instead of 27.5%.

By decreasing the transfer price, 20 of pre-tax income is shifted from Company B to Company C and taxed at 20% instead of Company B’s 30% rate. Because Company B and Company C are members of the same controlled group, they are largely indifferent as to which company earns the income, while the group as a whole pays lower overall tax (20 × 10% = 2). The after-tax return is increased from 52.5 (PBT of 80 – tax of 27.5) to 55 (PBT of 80 – tax of 25), or from 26.25% (52.5/200) to 27.5% (55/200), with a simple and non-transparent change. This example illustrates what may be thought of as “plain vanilla” transfer pricing.

Before turning to the next example, it must be acknowledged that this example is greatly simplified to illustrate the point. Many other tax and non-tax factors can come into play and constrain what taxpayers can charge. Customs duties and covenants in loans from third-party lenders are examples, but experience demonstrates that in almost all cases these constraints can be worked around when the tax savings are material.

As observed, in the international context it is more and more common, if not customary, for a multinational business to structure intercompany transactions using a company in a third country to act as an intermediary and carry out one or more elements of the transaction. An intermediary company often is used to accomplish tax reduction objectives. A multinational can search its global portfolio of companies to find a low-taxed entity to use as an intermediary, or if the amounts involved justify the cost, it can form a new entity for the transaction or structure.

Example 3

Assume the same facts as in Example 1, except that Corporation B establishes Subsidiary D in Country Z. Country Z taxes income at a rate of 5% and has no dividend withholding tax. Assume that Country X has a dividend exemption system so that Corporation B pays no tax on earnings distributed from Subsidiary D. Now assume that Corporation B sells the coal it mines to Corporation D for 120. Corporation D incurs 10 of marketing expense and sells the coal to Corporation C for 170. Company C incurs 10 of selling expense and sells the coal to customers for 200. The same pre-tax income of 80 now is divided, 20 to Company B, 20 to Company C and 40 to Company D. See Figure 3.3. The pre-tax income and taxes paid by Corporations B, C and D and the Company A group’s effective tax rate (ETR) are shown in Table 3.1.

Table 3.1Example 3: EBT and taxes by company
CountryCountry tax rate (percent)CorporationPre-tax incomeTax

Figure 3.3Low-tax intermediary – example 3

By adding flexibility to shift income to an even lower-tax country, the intermediary case offers greater opportunities for taxpayers and poses challenges for governments. By shifting 40 of income from Company B (taxed at 30%) to Company D taxed at 5%, 12 in tax is saved. The after-tax return is increased from 52.5 (PBT of 80 – tax of 27.5) to 68 (PBT of 80 – tax of 12), increasing the after-tax return on sales from 26.25% (52.5/200) to 34% (68/200). While not as simple and non-transparent a maneuver as in Example 2, if the Company A Group already had a Company D in Country Z carrying on other activity, the pricing structure would still be reasonably non-transparent.3

Starting from the base line of Example 1, these simple transfer pricing steps reduced the effective tax rate from 27.5% to 15% and increased the after-tax return on sales from 26.25% to 34%. It should be evident from these examples that transfer pricing has the potential to significantly affect government revenues. Analyzing the appropriateness of this transfer pricing requires a detailed understanding of the facts surrounding the transactions, making this planning difficult for governments to monitor. Moreover, this planning involves the largest and most sophisticated taxpayers, whose aggregate income is measured not just in millions but in billions of dollars (sometimes exceeding the GDP of the host country). The evidence of tax-motivated income shifting in the context of developed countries is strong (Clausing, 2011; Grubert, 2012). There is less available data from developing countries, but the available data and anecdotal evidence support a strong inference that the same problems arise in a developing-country context (ActionAid, 2012, 2013; Fuest and Reidel, 2010; Schatan, 2012). It is clear that transfer pricing is important to governments as a matter of revenue protection.

Another reason transfer pricing is important is that, if one taxpayer engages in non–arm’s length pricing and another does not, the differences in outcomes can result in unfair disparities between taxpayers. If in Example 3 the income shifted to Corporation D is not consistent with arm’s length pricing, it allows the Corporation A group to have a higher after-tax return (with an effective tax rate of 15%) that would advantage this group over another business that does not engage in non–arm’s length transfer pricing.

The potential for revenue loss and disparate treatment of competing taxpayers are not the only reasons transfer pricing is important. Non–arm’s length transfer pricing may distort real investment decisions as well as profit location.4 Moreover, the cost of planning to maximize tax savings from transfer pricing is not a socially productive use of resources. These costs are reasons in addition to revenue loss that transfer pricing is an important issue. It may not always be clear whether or the extent to which the host country benefits from tax competition, though recent work suggests it may be harmful (IMF, 2014, pp. 20–21).

Permitting tax avoidance through transfer pricing likely is not a desirable way to attract investment, because the tax loss is not related to the scope of the benefit to the host country. From the perspective of the host country, preservation of tax revenue justifies attention to transfer pricing issues (Schatan, 2012).

In summary, transfer pricing is the price charged for transfers of goods, services, property or other items of value in transactions between persons that are under common control. Three reasons transfer pricing is an important issue for any tax system are:

  • The potential for substantial revenue loss to governments from shifting income to countries with low or no income taxes

  • The potential unfair advantages for taxpayers that engage in non–arm’s length transfer pricing

  • The potential for distorting the allocation of activity among countries and possible economic welfare loss

In the developing-country context, the primary focus generally is on revenue loss, but that is not the only benefit from well-managed transfer pricing administration and enforcement.

3 What are taxpayers’ and governments’ transfer pricing objectives?

Taxpayers organize their business operations to achieve efficient delivery of goods and services to customers and at the same time to maximize global after-tax returns. Global businesses take advantage of modern communications to carry out functions either as part of the overall corporate group or through out-sourcing according to whichever approach is most cost-efficient. It is more feasible today than in the past for a company to exercise quality control as a purchaser or customer using contractual rights and remote as well as on-site monitoring. Businesses also use multiple geographies as necessary. The ability of business to separate functions, to carry on functions that earn a higher return and to subcontract other functions, each in multiple geographies, also brings with it the ability to minimize taxes.

Taxes increasingly are viewed by MNEs like any other cost, and tax structuring is layered into and integrated with the functional analysis described in the preceding paragraph. MNEs routinely establish structures that allow them to minimize taxation in the country where activity is performed and shift income to lower-tax countries. Sometimes the tax and non-tax planning go hand in hand. Where it is possible to achieve a favorable tax result without disrupting the needs of the business, MNEs will take steps to accomplish that result. Equity analysts and the business media review an MNE’s tax disclosure in its financial statement and take note of the company’s effective tax rate (taking into account whether the rate is sustainable).

There are important constraints on MNE tax planning. In relation to an income tax, taxes only come after there are profits from the business. It does not make sense to spend time and resources on tax mitigation unless the potential tax savings will exceed the direct and indirect costs of the planning. The indirect costs can be substantial if tax-related planning adds complexity to business operations (e.g., through frequent pricing adjustments) or uses valuable management time (such as requiring executives to attend board meetings in remote locations to satisfy “substance” requirements).5 If the tax savings are substantial enough, however, a company will take the actions necessary to support the tax position.

Transfer pricing is an integral part of an MNE’s overall tax planning. Many countries require documentation of material transfer pricing positions, and in some there are penalties for failure to produce the documentation (OECD, 2012, p. 20). Transfer pricing is an important item reported in financial statement tax reserves, and financial statement auditors reviewing a company’s tax reserve generally will review material transfer pricing exposures (IRS, 2015). An MNE generally seeks to avoid adjustments that increase its tax liability, but a publicly traded MNE also focuses on avoiding a tax adjustment that is not reserved against in financial statements and would result in a reduction in reported earnings. Accordingly, for corporate taxpayers there may be a helpful tension between seeking to reduce taxes through transfer pricing and seeking to avoid adjustments to financial statement earnings. This tension depends in part on a host country having the legal and practical ability to bring a transfer pricing adjustment with some likelihood of success. The capacity to administer transfer pricing rules and litigate, if necessary, adjustments to a successful conclusion is lacking in some, if not many, developing countries (IMF et al., 2011, p. 34).

The objectives of a government in applying transfer pricing rules principally are (i) to promote neutral tax treatment of controlled and uncontrolled transactions, (ii) to collect its fair share of revenue from national economic activity and resources and (iii) by following international transfer pricing standards, reasonably interpreted and applied, to signal to foreign investors a general intent to avoid international double taxation of the same income and to adhere to recognized standards in relation to taxation issues. Each of these objectives bears elaboration.

A fundamental objective of transfer pricing is to promote efficient economic decision making in the context of applying an income tax. Absent special circumstances, neutrality in taxation of domestic and cross-border businesses operating in the same country and industry should be welfare enhancing. It is good tax policy to minimize systemic tax advantages and disadvantages for businesses that compete, whether directly or indirectly, for economic resources within the same economy. As discussed, efficiency concerns may not carry the same policy weight in the context of a developing country’s taxation of extractive industries. The resource sector in developing countries generally involves limited competition between inbound foreign direct investment and domestic industry, though important exceptions apply in relation to artisanal mining and similar local industry.

In addition to the more abstract efficiency and fairness reasons to identify a “neutrality” objective, a perhaps more important reason is to preserve public support for the tax system as a whole. Neutral treatment of controlled and uncontrolled transactions is necessary so that multinational business is not perceived to have a tax advantage over domestic business. If multinational business is allowed a systemic advantage as a result of the country’s tax treatment of transfer pricing, there are multiple sources of risk to the tax system. One is that public outrage pushes for political solutions that are superficially appealing but are ineffective or counterproductive over the longer term.6 A second risk is that public cynicism sets in or is increased, such that the public will to comply with the tax law generally is eroded. Every tax system relies on public trust. There is a very high cost from losing the public’s confidence in a tax system’s fairness, including the cost and difficulty of restoring trust that has been lost.

A second objective is to further the primary purpose of a tax system, which is to collect the revenue authorized by law to fund public goods and services. One risk to tax revenue is the inappropriate allocation by a taxpayer of income to another fully taxing country. In this case, the taxpayer may be indifferent regarding the allocation of the income; the two countries should where possible try to agree on a consistent allocation of the income. This generally requires legal authorization under an international agreement between the two countries, alternatives for which are discussed later in the chapter. Increasingly, though, income is allocated to a low- or untaxed intermediary company in an intermediary country that makes no effective claim to tax the income. In this case, the risk is effective double non-taxation, and the principal beneficiary of the transfer pricing is the taxpayer not the other country. Because these cases typically involve taxpayer planning, they involve the greatest potential risk exposure and come up often in the context of cross-border extractive industry tax planning.

The third objective, to reasonably apply international transfer pricing standards, has as one important purpose signaling to potential foreign investment that reasonable transfer pricing will be respected. This does not imply renouncing responsibility to closely scrutinize controlled transactions under applicable law. It does imply being able to justify adjustments under a reasonable interpretation of international standards and trying to mitigate actual international double taxation (amounts paid to two countries on the same income). As discussed later in the chapter, reliability in taxation is particularly important to investors in the development phase of mineral extraction because of the long-term nature of the investment.

There is a natural tension among the government’s objectives between administering the tax law according to its purpose to collect revenue and the important objective for most developing countries of fostering a favorable climate for investment (the latter may be better addressed through tax design and non-tax measures). The tensions between the interests and objectives of governments and investing businesses are clearer in relation to, respectively, obtaining tax revenue and mitigating tax liability. In both cases, the disparate objectives should be reconciled by principled and even-handed application of the law and not with special arrangements or other non-transparent mechanisms that provide opportunity for rent-seeking payments and corrode public trust.

4 Why is the arm’s length principle the internationally agreed standard for transfer pricing?

The arm’s length principle is described in the OECD Transfer Pricing Guidelines by reference to the operative language of paragraph 1 of Article 9 of the OECD Model Tax Convention, which provides in part as follows:

[Where] conditions are made or imposed between the two enterprises in their commercial or financial relations that differ from those that would be made between independent enterprises, then any profits that, but for those conditions, would have accrued to one of the enterprises, but by reason of those conditions have not so accrued, may be included in the profits of that enterprise and taxed accordingly.

(OECD, 2010, ¶1.6)

The Guidelines go on to provide:

By seeking to adjust profits by reference to the conditions which would have obtained between independent enterprises in comparable transactions and comparable circumstances (i.e. in “comparable uncontrolled transactions”), the arm’s length principle follows the approach of treating the members of an MNE group as operating as separate entities rather than as inseparable parts of a single unified business.

(OECD, 2010, ¶1.6)

The arm’s length principle also is known as the separate accounting method or separate transactions method, since it treats commonly controlled entities as separate and evaluates the transactions between the members of the group. The base line reference point for transfer pricing is what an independent enterprise would have charged in comparable transactions and comparable circumstances. What does this mean?

If the touchstone of the arm’s length method is comparability with an arm’s length transaction, how does a tax authority or a taxpayer identify a comparable transaction when for the transaction in question the taxpayer decided to contract with a related instead of an unrelated person? And what is to be done if there is no unique price that would be reached by independent parties (e.g., because of different bilateral negotiating power)? Under what “comparable circumstances” would there be a “comparable transaction’ with an unrelated person?

The paradigmatic comparable transaction is found when the same taxpayer is selling the same property to an unrelated person as to the related person in the same market (under the same terms of sale). This is not a frequent circumstance.7 Sales of commodity-type fungible goods may provide a market yardstick against which to measure a related party sale. However, even in those cases marginal under valuations can have a material impact in large-volume or large-value transactions. The difficulty inherent in the arm’s length principle is applying it in practice (Schatan, 2012, p. 126). Determining an arm’s length price requires knowing the specific facts of the transaction and understanding the market in which the transaction occurs. The transaction-based and profit-based methods set out in guidelines are described in what follows, but in all but the simplest cases arm’s length pricing involves market knowledge and reasoned judgment.

The principal alternative to the separate accounting approach is to take the profit of a business conducted in multiple jurisdictions and allocate it among the jurisdictions in which the business is conducted according to a formula that uses proxy measures intended to identify where the income is earned or which jurisdiction should have the claim to tax profit. Although formulary apportionment methods have not been used at the national jurisdiction level, many states of the United States have employed formula apportionment methods. In addition, formulas are used to apportion income among Canadian provinces and Swiss cantons and the EU Commission has proposed a common corporate consolidated tax base (CCCTB) allowing use of formulary allocation of income to member countries (EU Commission, 2011; Hellerstein, 2012, p. 223). The typical factors are property, payroll and sales, but experience shows that jurisdictions adjust these factors to take account of their own circumstances for their own advantage. There has been a robust debate over many years about the relative advantages and disadvantages of formula apportionment in relation to separate accounting arm’s length pricing as a method for allocating international income, but it so far has not been adopted as an international standard.

Why is arm’s length the international standard? The answer is found partly in the objective that transfer pricing regulation is trying to achieve and partly in history and path dependency. The more powerful reason is the primary objective of the transfer pricing rules: to achieve broad parity of treatment for associated enterprises and independent parties. In other words, transfer pricing rules should be aimed at achieving tax neutrality in the decision whether to engage in a controlled transaction or an uncontrolled transaction (Grubert, 2005, p. 149; OECD, 2010, ¶1.08). Ideally, the optimal after-tax decision regarding choice of a controlled versus an uncontrolled transaction is the same as the pre-tax decision. Similarly, the choice of how to conduct and where to locate economic activity should be unaffected by considerations of transfer pricing.

There is substantial indirect evidence that this objective is not always met in developed as well as in developing countries. The indirect evidence comes in the form of aggregate data showing foreign margins increasing in inverse relation to the foreign tax rate and disproportionate income in low-tax countries (Clausing, 2011; Grubert, 2012). In public hearings, some prompted by media reports, indirect anecdotal evidence of income shifting by individual companies has been brought into public view (U.S. Senate [PSI], 2012, 2013; UK House of Commons [Public Accounts], 2012). In addition, activist groups and media have highlighted cases of individual companies in developing countries using transfer pricing along with other tax planning techniques to shift income from developing countries to havens or low-taxing countries (ActionAid 2012, 2013; Bergin, 2012a; Schatan, 2012).

In light of this accumulating evidence, why would countries continue to follow the arm’s length principle in transfer pricing, particularly if income is lost to havens and low-tax intermediary countries? One question is whether formulary apportionment would be better, particularly for developing countries with natural resources. When the usual apportionment factors of property, payroll and sales are considered, it is not likely that resource-rich developing countries would fare well. Payroll typically is low. Most sales are to foreign buyers, and these sales may or may not be shifted outside of the country. Finally, the extractive industry companies generally do not own the mineral property but operate by concession, so the property factor likely would be modest. Moreover, even if the resource property investment were taken into account, property values do not reflect the discovery value of the minerals. An investment at cost understates the value of a successful find and development. There is substantial reason to think formulary apportionment would not be in the interests of a developing country with an important extractive industry sector.8

Another powerful reason not to move to formulary apportionment is simply the fact that the international tax regime has achieved a remarkable consensus around the arm’s length principle, and it is extremely unlikely that a comparable level of consistency could be achieved in relation to any form of apportionment in a reasonably near time frame (Fleming, Peroni and S. Shay, 2014, p. 9). Consistency is a key, since inconsistent income allocations can result in double non-taxation, which results in windfall gains for affected taxpayers and revenue loss to governments that has to be made up with higher taxes on other taxpayers and double taxation, which discourages investment (OECD, 2015c).

There are ways in which the stress on transfer pricing can be reduced, and there are opportunities to improve enforcement through constructive engagement with investing companies. A range of potentially effective unilateral as well as multilateral strategies may be employed by governments to respond to transfer pricing challenges that are consistent with the arm’s length principle. The starting point is risk assessment, structural responses and thoughtful choice of cases for enforcement. A taxpayer’s incentive to engage in income shifting is related to the costs to structure, implement and defend the arrangements, so there has to be enough benefit from aggressive planning to justify the direct and indirect costs. Tax law changes can materially affect taxpayer incentives. Enforcement matters, though governments with scarce enforcement resources must target enforcement where the payoff is highest. Possible host government responses to address transfer pricing are discussed in Section 6.

5 How is the arm’s length principle applied?

5.1 The OECD guidelines and UN manual approaches

The OECD Transfer Pricing Guidelines and the United Nations Practical Manual on Transfer Pricing each set out generally accepted methods for determining a transfer price that may be appropriate in circumstances in which relevant comparable uncontrolled pricing information is available to allow the method to be applied in a way that is likely to yield a reliable arm’s length price. The UN Manual acknowledges that

in reality two transactions are seldom completely alike and in this imperfect world, perfect comparables are often not available. . . . To be comparable . . . means that either none of the differences between them could materially affect the arm’s length price or profit or, where such material differences exist, that reasonably accurate adjustments can be made to eliminate their effect.

(UN, 2013, ¶5.1.5)

This captures the practical reality and difficulty of transfer pricing.

The UN Manual describes arriving at an arm’s length price as involving a process that includes determining the “economically significant characteristics of the industry, taxpayer’s business and controlled transactions,” identifying comparable transactions and making comparability adjustments as necessary and applying the most appropriate transfer pricing method to determine an arm’s length price or profit (or range of prices or profits; UN, 2013, ¶5.2). The following discussion follows the UN Manual terminology and structure, which does not differ in any material respect from that of the OECD Transfer Pricing Guidelines. The five major transfer pricing methods are broken into two groups of methods, three so-called “transaction-based methods” and two profit-based methods.” The UN Manual describes the transaction-based methods as follows:

Transaction-based methods

1.5.4. Comparable Uncontrolled Price (CUP) The CUP Method compares the price charged for a property or service transferred in a controlled transaction to the price charged for a comparable property or service transferred in a comparable uncontrolled transaction in comparable circumstances.

1.5.5. Resale Price Method (RPM) The Resale Price Method is used to determine the price to be paid by a reseller for a product purchased from an associated enterprise and resold to an independent enterprise. The purchase price is set so that the margin earned by the reseller is sufficient to allow it to cover its selling and operating expenses and make an appropriate profit.

1.5.6. Cost Plus (C + or CP) The Cost Plus Method is used to determine the appropriate price to be charged by a supplier of property or services to a related purchaser. The price is determined by adding to costs incurred by the supplier an appropriate gross margin so that the supplier will make an appropriate profit in the light of market conditions and functions performed.

(UN, 2013, ¶¶1.5.4–1.5.6)

In these transaction-based methods, a search must be made for a relevant comparable transaction, comparable resale price margin or comparable gross margin on costs, respectively, to reach operating profit. The transaction-based methods may be a reliable indicator of a market price where a comparable transaction may be identified; however, as observed in the UN Manual quote above, a comparable transaction often is unavailable or requires too many adjustments to be considered reliable. If transaction-based methods are not appropriate for the tested transaction or relevant comparable transactions are not available or are not reliable, then another alternative is to use “profit based-methods,” which are broken into profit-comparison methods and profit-split methods.

The profit comparison methods are called the “transactional net margin method” (TNMM) in the OECD Guidelines and the UN Manual and, alternatively, in some countries such as the United States, the comparable profits method (CPM). (As a practical matter, these methods are interchangeable.) The profit comparison methods involve finding businesses that are functionally similar to the business whose related party transactions are being tested but whose operating margins are not affected by related party transactions (and may be found in the public record). The comparable businesses’ margins are used as the basis to identify a profit-level indicator metric (such as an operating return on invested capital) that is applied to the relevant part of the tested business to determine an arm’s length return. The profit split methods similarly look to unrelated businesses for an allocation key that may be used to test the division of profits in the related party case.

The profit-based methods are described in the UN Manual as follows:

Profit-based methods

1.5.8. Profit comparison methods (TNMM/CPM) These methods seek to determine the level of profits that would have resulted from controlled transactions by reference to the return realised by the comparable independent enterprise. The TNNM determines the net profit margin relative to an appropriate base realised from the controlled transactions by reference to the net profit margin relative to the same appropriate base realised from uncontrolled transactions.

1.5.9. Profit-split methods Profit-split methods take the combined profits earned by two related parties from one or a series of transactions and then divide those profits using an economically valid defined basis that aims at replicating the division of profits that would have been anticipated in an agreement made at arm’s length. Arm’s length pricing is therefore derived for both parties by working back from profit to price.

(UN, 2013, ¶¶1.5.8–1.5.9)

The metrics for dividing profits, particularly under the profit split method, places these methods somewhere between transaction-based methods and a formulary apportionment approach. One key distinction is the extent to which an attempt is made to tie back to market-based benchmarks in developing a profit-level indicator or allocation key. The comparable profit benchmarks usually are obtained from searches of databases and yield a range of profits. M. Durst (2010) observes that multinational businesses that distribute products, provide services to affiliates or manufacture in developing countries or some combination of those activities “overwhelmingly use ‘profit-based’ pricing methods.”9 Another key distinction is that formulary apportionment allocates the net income of an enterprise or line of business, whereas the arm’s length methods determine the price for a transaction or grouping of similar transactions.

Modern transfer pricing rules accept that in many if not most cases there is not a single arms’ length price for a transaction but that prices within a range of prices may be accepted as arms’ length (UN, 2013, Ch. 6). This recognizes the inexact nature of the transfer pricing exercise and forestalls minor adjustments. But even allowance of a range of outcomes does not surmount fundamental difficulties in applying the arms’ length standard in the absence of a market comparable (Auerbach, Devereux and Simpson, 2010; Benshalom, 2013). There are numerous circumstances in which an action would be taken between related parties that would not between unrelated parties; the arm’s length standard still must be applied, but it cannot be by reference solely to an inapposite comparison with a transaction that occurs between unrelated parties. Instead, it is necessary to ask what rational economic actors would do in the circumstances (OECD, 2015c).

In addition to conceptual difficulties, the fundamental differences between the information available to the taxpayer and that available to the government make enforcement that achieves realistic outcomes extremely difficult (Shay, 2009, pp. 327–330). This provides particular advantage to taxpayers that use a “one-sided method,” that is, a method that only looks to the results of one of the related parties to the transaction. (The transfer pricing methods identified other than the profit-split method only analyze one side of a related party transaction.) A one-sided approach can hide outcomes that if made fully transparent would be subject to question. Use of a “two-sided” profit split method, whether as a primary method or as a test of the results of another method, protects against outlier outcomes in the related counterparty. Nonetheless, taxpayers resist disclosing information about the party outside of the jurisdiction and providing information to determine the “see through” profit for a line of business. In the author’s experience, this information often is critical to fairly evaluate the allocation of income.

Fee charges for management and other “high-value” services are an ongoing and material risk exposure for developing countries. At the other end of the risk spectrum are charges for sharing of routine or ministerial services. Some countries have cost-based transfer pricing methods for routine services. The OECD observes that there are cases of low-hanging fruit for transfer pricing enforcement: “if a company is paying for goods and services (such as Head Office costs) when it has not received any real value, the cost may be disallowable on first principles” (OECD, 2012, p. 69). Even in relation to cost-based services, when a pool of costs is charged out, it is important to confirm that the costs have actually been incurred and that the allocation key is reasonable and appropriate (Schatan, 2012).

This chapter will not further discuss the specific methods recommended for determining an arm’s length price in relation to the various categories of income from sales of tangible goods, performance of services, transfers of intangibles and so on and how to determine comparability of uncontrolled transactions and data for profit based methods. These are extensively, indeed exhaustively, described in the OECD Guidelines and the UN Manual (OECD, 2010, 2015c; UN, 2013).

While familiarity with transfer pricing methods and comparability guidelines published by the OECD and the United Nations provides a framework for consideration of transfer pricing in general terms, these documents are not directed at extractive industry transfer pricing and are of limited help in identifying risk exposures in extractive industries. The OECD Global Forum on Transfer Pricing recently published a draft Handbook on Transfer Pricing Risk Assessment. This is a promising tool for risk assessment; however, it also does not address extractive industry issues except in passing. The sole reference to extractive industries is in paragraph 38, which gives this example of a recurring transaction risk exposure: “For example, if a local taxpayer in one of the extraction industries sells all of its local country output to related entities, small pricing discrepancies in each individual sale can add up to large reductions in the local tax base” (OECD, 2013c, p. 38). This provides an appropriate segue to the discussion of extractive industry transfer pricing challenges.

5.2 Extractive industry transfer pricing challenges: determining the sales price

In many cases, a developing country with an important extractive industry will rely on the revenues from the extractive industry for a material portion of its government revenues. In a review of 57 resource-rich countries, oil and gas revenues in 22 petroleum-producing countries surveyed by the IMF rose to a weighted average of 35% of government revenues in 2010. Revenues from mining were more volatile, but in the period 2006 to 2010 ranged from 10% to over 25% of government revenues (IMF, 2012, p. 62).

Oil and gas and mining share some important characteristics and differ in others. Both industries require substantial up-front capital investment that is not easily recoverable until sustainable production is achieved. There is substantial risk of market price variability, and cost recovery extends over long periods. Individual projects will have a finite life. Generally, exploration is riskier in oil and gas, but development of the resource and exploitation is riskier in mining (IMF, 2012, pp. 8–10).

Natural resource exploration and development typically are governed by a concession agreement that governs the obligations of the company or consortium undertaking exploration and development and the host government (CCSI, 2015). Most such agreements include provisions that attempt to freeze the law governing taxation of income from the project for the life of the project (so-called stabilization clauses) and the dispute resolution clause may or may not pre-empt the usual host country process for resolving tax claims. The agreements’ royalty and tax provisions normally refer to fair market value for selling prices but typically do not address or otherwise limit application of normal transfer pricing rules in the determination of selling price for income tax or royalty purposes.

It is common under a mining or petroleum concession for a company exploiting mineral resources (particularly after the exploration stage) to be a multinational enterprise with substantial production and marketing skills and a broad geographic footprint (IMF, 2012, p. 8). In many, if not most, such cases, the initial sale of the mineral is to a related party. In cases in which the affiliate buyer is engaged in further processing or other substantive activity in relation to the mineral, as opposed to serving as a conduit selling immediately to an unrelated buyer, these inter-affiliate sales present important transfer pricing risks and challenges.10

The following discussion considers what would appear to be a simple question, how to determine the arm’s length sales price of a controlled sale of a mineral. For purposes of this discussion, the mineral will be iron ore.

It is customary in a mineral-development arrangement for the host country to charge a royalty on the sale of the mineral at the port of shipment as well as impose income tax on the profit of the producing company (IMF, 2012, pp. 30–33).11 The sales price will affect, possibly materially, the royalty charged on gross sales of the mineral, which is received by the host country once the mineral is first shipped, as well as the gross income for determining the corporate income tax, which is realized by the host country after the mining company’s income exceeds its period costs and, in some countries an additional profit or “resource rent tax.”12 Over the life of a mining project, no single issue is more important for determining the host country revenue from the project than the price at which the mineral is sold.13

To the uninitiated, there may be a view that determining the sales price for a mineral is straightforward and does not pose meaningful transfer pricing risk. Are there not published indices? Unfortunately, the issue is not so simple, and the circumstances can differ materially according to the mineral in question. Where the sales price is intended to compensate the producer solely for production and not for manufacturing into ingot, pricing for income tax purposes usually is at the port of shipment for minerals (see Chapter 4 by Calder, this volume). Determining the sales price for minerals, such as iron ore sold “FOB port,” can have practical complications, as discussed in what follows.

Each shipment of iron ore has differences in quality. Iron ore has impurities, including silica and alumina, which affect the efficiency of a blast furnace in processing the ore. The silica and alumina content can vary for ore from the same mine, depending on where in the mine the ore is taken, as well as for ore from different mines. In addition to impurities, the moisture content of the ore affects the freight and processing. If the moisture content is too great, the ore simply cannot be shipped by ocean vessel because of the danger of moisture affecting the stability of the ship in certain conditions. Insurers will not insure cargos outside certain parameters for moisture (Chapter 4 by Calder, this volume).

The explosive growth of Chinese steel manufacturing changed the dynamics of the iron ore industry and has affected the pricing mechanism for iron ore. Pricing has moved from changes made once a year to pricing for each shipment, generally at an average of market prices for a recent period such as a month. Since 2010, industry indices, such as Platts IODEX, have provided daily China cost and freight (CFR) prices,14 which means that FOB port prices used for royalties and income tax payments to a producing country must be net back from the index CFR China prices.15 The Platts index is constructed from a survey of prices paid by purchasers in the relevant market for ore within its specifications (Platts, 2015). It is common to use the Platts iron ore index price for the iron ore grade closest in iron (Fe) content to what the mine produces and adjust from that price for each shipment. Key variables in arriving at an arm’s length FOB port price for an iron ore shipment in a controlled transaction include:

  • Which Platts iron ore grade (i.e., 62% Fe, 58% Fe, 52% Fe, etc.) will be used as the base from which to adjust for Fe content? Each grade has prescribed tolerances for silica and alumina, though the actual tolerances accommodated by producers may vary.

  • Will the Fe content adjustment be based on straight-line between prices for surrounding grades or take account of the increasing value of higher Fe content?

  • What will be the basis for adjusting from the index for impurities that exceed the tolerances for the grade used as the measuring point?

  • Will freight be based on actual freight or, if the taxpayer has available a portfolio of ships, be based on an index to avoid risk of adverse ship selection?

  • If the taxpayer sells ore shipments to different steel producers, should the CFR priced be reduced for a marketing allowance to the intermediary entity?16

  • If the mine is a new mine, should a trial allowance discount be allowed to encourage different steel producers to use the ore in their furnaces?

As is evident from the technical nature of the issues described and further taking account of the fact that iron ore pricing methodologies are in a period of flux, a host country should work with an industry expert to establish a method for determining an arm’s length FOB port price for iron ore from a major concession. An MNE’s knowledge of its industry and its own business poses a challenge for a tax authority in relation to any industry, but this is especially true for extractive industry issues. It is a powerful illustration of the breadth of the so-called “information asymmetry” advantage that MNEs have in transfer pricing generally.17

From a transfer pricing perspective, the iron ore case is an example of using a comparable uncontrolled price, based on an index, and making adjustments to reach an “arm’s length price” or fair market value. Concerns for the host government’s administration of transfer pricing that may be identified to this point are (i) the potential for disparity in technical knowledge of the industry and market between a global multinational in a specialized industry and the host government and (ii) the host government’s need for information regarding the multinational’s operations relevant to the mining business. Even limiting consideration to the sale price of the property, it would be valuable for the host government to have a “see through” profit analysis of the iron ore business and test a derived comparable price by seeing the resulting profit split from the business with all of the relevant affiliates. This would be using a “two-sided” method to test the results under a one-sided transactional method. Finally, it seems very unlikely, even acknowledging potential weaknesses in using the Platts index as a starting benchmark price, that a host country would be better off on almost any basis applying a formula apportionment regime to this or indeed to most any other mineral extraction case.

While the pricing of the extracted product is the most significant potential transfer pricing issue, even if a host country is successful in achieving an arm’s length price for the physical product, there are numerous other ways that an MNE can reduce its effective rate of tax in the host country that also have transfer pricing elements. The next subsection considers other, systemic transfer pricing challenges that a host country for an extractive industry investment must face. These challenges are common across a range of industries engaged in substantial foreign direct investment in host countries, including non-extractive industries such as telecommunications.

5.3 Extractive industry transfer pricing challenges: a host country tax minimization example

Once the sales income of the in-country producing company is determined, the objective of an MNE will be to achieve the lowest practical global tax rate taking account of the transactions cost of engaging in tax planning. Another element in a transfer pricing review should be to assess whether the MNE investment in its plant and equipment includes material controlled purchases that should be assessed. Use of “purchasing company” markups is a common way to shift profits, but it will not be discussed further here (Tuerff et al., 2011). Other standard income shifting techniques that are routinely used to reduce the effective tax rate of a producing company include (i) holding the company through a favorable low-tax country, such as the Netherlands, Mauritius or some other location, (ii) capitalizing the producing company with intercompany debt, (iii) charging management and other service fees to the producing company and, in some cases, (iv) charging the producing company with a royalty for use of technology, know-how, a trademark or trade name, as the case may be.18

The offshore holding company or another offshore affiliate often will fund the producing company with material amounts of intercompany debt, which (subject to third-party loan covenants) permits repayments of principal, which are not subject to the restrictions on dividends (such as a requirement that the company have current or retained earnings or surplus). In addition, interest on the intercompany debt generally is deductible, sometimes even when accrued but not paid with cash (but by issuing additional debt or increasing the principal amount of the debt).19 Often the offshore holding company (or another affiliate) also will charge a producing company a management services fee. While in some cases this is passing through allocated costs of a central or regional headquarters, the fees also may be marked up with a profit margin. As noted, using intangibles to strip profits also is an additional strategy. These strategies are illustrated with respect to interest and management fees in Example 4.

Example 4

Assume that an iron ore producer’s parent company is Company A organized in Country W. It wholly owns an offshore holding company, Company B, organized in Country X, a low-tax jurisdiction. Company B owns all of the stock of the iron ore producer, Company C, in Country Y where the mine is located. The mine is the largest source of potential mineral revenues in Country Y. Company C sells most of its output to steel producers in Asia or Europe. On each continent, Company C may sell ore to a local distribution company, Company D in Country Z, wholly owned by Company B, to Company B for on-sale to customers or directly to customers. Company B reduces Company C’s Country Y tax through charges for interest on intercompany debt and management fees. See Figure 3.4.

Figure 3.4Host country tax minimization – example 4

Assume that Company C has 500 of sales income and 100 of income before taxes, interest and fees. If the Country Y tax rate is 30%, Company C will pay 30 in tax to Country Y. If Company B makes an intercompany loan to Company C of 100 at 13% interest and charges Company C a management fee of 20, Company C’s Country Y income before tax will be 67 and its Country Y tax will be 20 instead of 30. Assuming no Country Y withholding taxes by reason of treaties or a concession agreement and a Country X tax rate of 10%, the Group’s Country X tax will increase by approximately 3 for a net tax reduction of 7. The Group’s effective tax rate on the mining income is thereby decreased from 30% to 23%.

Several pricing issues are raised by charging service costs: were the costs actually incurred, do the allocated costs benefit the producing company’s business such that they should be allowable as a deduction by the host country, are the costs allocated under a reasonable methodology consistently applied to all similarly situated affiliates over relevant years and is the profit markup (if any) arm’s length? With respect to interest, the questions include: Is the amount of debt and the interest charged on the debt arm’s length in amount (OECD, 2015d)? Issues similar to those for service fees and royalties also apply to royalties (Schatan, 2012). These issues are susceptible to audit, but it also is possible to consider tax law changes that would reduce the potential benefits from this kind of intermediary tax planning. In this respect, tax system design, within a corporate income tax, can affect incentives to engage in transfer pricing as a tax avoidance strategy.

This leads to the question, what strategies are there for governments to follow to address transfer pricing exposures and the base erosion strategies described in Example 4?

6 How government strategies respond to transfer pricing challenges

1 Have a strategy

Is the situation with transfer pricing hopeless? Of course it is not. Like any other tax issue, transfer pricing compliance improves when rules are reasonably clear, incentives for aggressive transfer pricing are reduced and enforcement is focused, tough and fair. This section discusses several elements that should be included as part of an overall strategy to achieve the objectives described: protection of the host country’s tax base against profit shifting to offshore havens, reasonable application of international standards to avoid double non-taxation and double taxation of income and neutral treatment of controlled and uncontrolled transactions (so transfer pricing does not advantage MNEs).

6.2 Limit benefits from transfer pricing

As discussed, the incentives for transfer pricing are related to the extent to which taxes may be reduced. Under current international rules, it is difficult for a host country acting alone to affect the taxation of income earned from sales of minerals to persons outside the jurisdiction. While it is possible in theory to adjust the overall corporate tax rate, because of the revenue cost it is not practically possible to match the zero or very low rates of havens or low-tax countries. A host country can, however, scrutinize the transfer price. It can require the taxpayer to provide see-through profit information and test the price against the reasonableness of the resulting profit split among relevant affiliates in relation to the actual activity carried on by those affiliates. This should be made easier by adoption of country-by-country reporting called for by the OECD’s BEPS project (OECD, 2015e).

In the case of deductible payments or payments that may potentially be subject to withholding tax, however, many countries do limit the advantages of income stripping through imposition of a withholding tax on royalties, management and other service fees and/or by limiting deductions in relation to such payments as well as interest. It is important that these protections not be given up in concession agreements or income tax treaty provisions without obtaining equally meaningful economic benefits in exchange. A country should be able to protect its source tax base by measures that may include imposing withholding tax on and/or restricting deductions for deductible payments of income paid to or treated as beneficially owned by related foreign persons in countries that do not “effectively tax” the income. In addition to applying to hybrid mismatch arrangements, such a rule also should apply with respect to income paid to an intermediary entity where the income is subject to a low effective tax rate in relation to the source country’s tax on the payor company. The rate of tax on the recipient to apply for this purpose would be determined in relation to the level of tax that otherwise would apply in the source country (Fleming, Peroni and S. Shay, 2015; Lodin, 2013; OECD, 2015a).

In many cases, an intermediary company is not located in a treaty jurisdiction or, as in the case of a nonresident Irish company, is not eligible for treaty benefits. Where the intermediary is resident in a treaty country and is eligible for treaty benefits, however, the issue may arise whether in the particular circumstances the nondiscrimination limitation of Article 24(4) of the OECD and UN Model treaties would apply. In general, it appears that the OECD takes the view that nondiscrimination does not prevent application of properly designed anti-tax arbitrage rules (Fleming et al, 2015, pp. 704–709; OECD, 2015a, pp. 148–149). It is fundamental to a coherent international tax system that the discrimination principle be applied in a way that allows a country reasonably to protect its tax base (Graetz and Warren, 2006).

If the income of an intermediary company is taxed by a third country under a CFC regime, there is a risk that income would be taxed by both the host country and by the ultimate residence country in the same year. The OECD proposes that in a hybrid instrument case where a deductible payment is not included in income (D/NI), the primary response should be for the source country to deny the deduction. If the host country and the ultimate residence country in relation to an intermediary are parties to a bilateral income tax treaty, they could agree on a bilateral approach.

The objective of these changes would be to restrict the incentive to engage in the kind of planning and pricing issues that are raised in Example 4. A host government will need to evaluate the advantage of adopting provisions that protect its tax base in relation to a range of considerations that include investor perceptions. Moreover, even if base erosion provisions are added to a host country’s law, depending on how stabilization provisions in mineral concession agreements are drafted, in some cases the base erosion law changes may apply only to new concessions. In any event, the G20 and OECD attention to the base erosion issue is extremely helpful in validating the legitimacy of host country concerns and responses.

6.3 Transfer pricing legislation and regulations

A threshold question for many developing countries is whether they need transfer pricing legislation. In order to address transfer pricing abuses, it is necessary for host country tax authorities to have authority in their domestic law to make adjustments to income or deductions in controlled transactions. Most countries, however, do have provisions that grant substantial authority to the tax authorities to adjust a taxpayer’s income and deductions in transactions with associated persons (Tanzania, 2006). The question posed is more often a question whether in addition to broad statutory authority it is necessary to have more detailed legislation or regulations. Almost by definition, it is not “necessary” unless there is a specific reason to do so.20 In this regard, it is useful to distinguish between adoption of elaborate transfer pricing regulations and guidance that either address a specific industry context or is important for the tax authority to be able to administer and enforce arm’s length transfer pricing (Chapter 4 by Calder, this volume).

Whether to favor elaboration of detailed transfer pricing rules depends on the context. If a tax authority is willing to use broad statutory discretion, perhaps supported by guidance on principles, the courts (or an arbitrator) will support reasonable exercise of that discretion, taxpayers are deterred by the prospect of exercise of the authority and investors do not have reason to lose confidence in the reasonable exercise of the authority, little more may be needed. The United States and the OECD have promulgated mind-numbingly detailed rules hundreds of pages in length in the form of regulations (in the case of the United States) and guidelines in the case of the OECD (OECD, 2010; UN, 2013). The evidence from the United States is that highly detailed regulations have not staunched a virtual flood of profit shifting. More limited data available for other OECD countries is not inconsistent.

A possible reason to elaborate transfer pricing rules is to counter the possible effect of another country’s adopting more specific rules (and presumably stronger enforcement), such that there is concern that taxpayers perceive the path of least resistance is to allocate income to the other country (UN, 2013, ¶1.3.8). Effective enforcement of existing rules may be of greater relevance to address this concern, but it can be (and has been) a factor in countries’ decisions to adopt regulations.

Another potential reason to elaborate transfer pricing rules is to provide a “climate of certainty” to foreign investors. While it is more likely that a “climate of certainty” is determined by the level of confidence that investors have in the host country’s legal system and tax administration, the availability of rules to point to as a reason for a court to not make an adjustment also may support taxpayer confidence that a totally unreasonable position will not be sustained. In the particular context of extractive industry concessions, which generally include separate dispute resolution mechanisms (usually arbitration) outside of the host country court systems, it would be expected that taxpayers would have fewer concerns about the need to specify transfer pricing rules.

As mentioned, the OECD has promulgated detailed guidelines, and it is possible to adopt more generalized rules that make reference to those guidelines (without incorporating the OECD rules). A significant concern for tax authorities, especially those relatively inexperienced in transfer pricing, is that taxpayers may use any of the relatively general statements in the OECD Guidelines to the disadvantage of the tax authority in ways that will not advance the search for an arm’s length price. The OECD has addressed potential problems of excessively detailed rules with the publication by the Secretariat in 2010 of a paper that presents a simplified approach to drafting what could be either legislation or regulations (OECD, 2011). The draft proposed by the OECD includes rules for taxpayer documentation of their transfer pricing and authorizes correlative adjustments if an adjustment is made. While the paper is very helpful, it is advisable for legislation or regulations following its guidance and that make reference to the OECD Guidelines to make clear that the local law takes precedence over the Guidelines (Uganda, 2011).

6.4 Penalties, return information and documentation

An objective of a transfer pricing legal regime is to encourage compliant behavior from the first filing of the return and obviate the need for enforcement. The other side of “lowering the benefits” of engaging in aggressive transfer pricing is to increase its risks through adoption and enforcement of a transfer pricing penalty regime.21 The objective of the penalty regime should be to encourage taxpayers to take reasonable transfer pricing positions on their tax return by changing the risk calculus. One approach, which is used in the United States (but which is not well enforced), causes the penalty to increase as the size of the adjustment relative to the original tax return position increases. Thus, if a transfer price used on a tax return is 200% or more (or 50% or less) of the amount finally determined to be the arm’s length price, the penalty is 20%, but if the erroneous price is 400% or more (or 25% or less) of the amount finally determined, the penalty increases to 40%. In some countries, transfer pricing documentation serves as a defense against penalties.22

In transfer pricing, the critical advantage for taxpayers and problem for governments is the non-transparency of pricing issues generally. Unless specific issues are highlighted in lines or schedules of a tax return, there is little to signal tax authorities that there is a transfer pricing issue or what the issue involves. In some countries, the type and volume of controlled transactions is reported on a schedule to a tax return, which can be a useful screening tool to determine whether a transfer pricing exposure exists. Transfer pricing adjustments typically must be initiated by the government and therefore only are found on audit.

Another tool to help tax administrations assess transfer pricing exposures that has been increasingly adopted since the mid-1990s has been a “contemporaneous” documentation requirement that taxpayers demonstrate that their transfer pricing is arm’s length using the best method available and yields an arm’s length price.23 The utility of documentation as well as whether its value exceeds the costs of producing it depends on its design and the uses to which it is put. In some countries, preparation of documentation has become routine to the point that little thought is put into the analysis, and it is little more than a fig leaf to protect against penalties. Nonetheless, to the extent that it collects in one place information relevant to a pricing transaction and forces the taxpayer to take a position on the taxpayer’s pricing it has added some discipline to a process where previously there was none. It is a helpful starting point for a tax authority’s analysis of the transaction covered, but great caution should be taken in relying on public comparables used in documentation, as an array of techniques may be used to bias the sample of comparables with consequent effect on the soundness of the analysis.

If the taxpayer is unable to satisfactorily support its transfer pricing, then the tax authority may and should make an adjustment to restore the pricing to an arm’s length amount or, in the absence of a comparable transaction, a price that is within a range that rational actors would agree to under the actual circumstances of the transaction. It also is appropriate to assert penalties.24 If the taxpayer does not agree to the adjustment, the next stage is dispute resolution.

6.5 Transfer pricing dispute resolution

The objective of a dispute resolution mechanism is to achieve timely and cost-effective case resolutions. Because of the intensely factual nature of a transfer pricing case, litigation of transfer pricing cases in courts has in almost every case proven extremely time consuming and expensive for both governments and taxpayers. The resource intensity and delay of litigation has led countries to consider alternatives. Alternative dispute resolution mechanisms may be mutually advantageous for an MNE and a host government.

6.5.1 Advance pricing agreements

One alternative is for the government and taxpayer to reach agreement in advance on a method for determining pricing (an “advance pricing agreement” or APA). In a typical APA program, a taxpayer submits an APA request and discloses substantial information about the proposed transaction or business operation with respect to which a pricing agreement is requested. One question for governments is what standards it should apply in agreeing to consider a request. The overwhelming tendency has been to accept all requests, but that in essence cedes the allocation of APA resources to the decision of taxpayers as to whether to ask for advance guidance. It is reasonable to assume that taxpayers do not make requests in cases in which it may be advantageous to take a more aggressive position on a tax return and then defend it if attacked. Before accepting an APA case, the tax authority should have clarity about the objectives it hopes to achieve with the investment of resources in an APA for a single taxpayer.

One advantage of an APA for a government is that it can be more efficient than litigating a case. It may be questioned, however, whether the same taxpayer that voluntarily submits to the APA process would be adopting an aggressive pricing position in the alternative. Another advantage for a government is that it can learn about an industry from processing an APA. A taxpayer’s objective in reaching an APA is to achieve certainty. This may be particularly valuable in a large, long-term project such as an investment in mining production or oil and gas development. This taxpayer’s objective should not be underestimated as leverage for the government to achieve a favorable outcome in the particular case.

Another issue relates to transparency and disclosure of APAs. The government will benefit if an outcome is made public so that the result of its investment of resources is available to other similarly situated taxpayers. Transparency also reduces the risks of unequal treatment of taxpayers. Taxpayers typically are reluctant to disclose information about APAs, and many countries have restrictions on disclosure of taxpayer information. If there is not meaningful disclosure of the results of an APA, then the government is devoting resources to a single taxpayer with limited collateral benefits. If a government pursues an APA program, it should be thoughtful how to structure it so that it maximizes the use of its scarce resources. While the APA process has been used in a number of countries, the preceding discussion suggests why it is not a panacea for transfer pricing and certainly does not serve as a substitute for a robust enforcement program.

In the extractive industry context, an APA–type agreement might be considered as an integral part of negotiating the concession agreement and be required for all concessions in similar categories. This would enhance the bargaining leverage of the host country and could build in flexibility for changes that might be carved out from an arbitration procedure.25

6.5.2 Treaty mutual agreement procedures

As observed earlier, Article 9 of the OECD and UN Model Treaties provide the framework for arm’s length pricing. This standard has been incorporated in one form or another into most countries’ domestic laws, so arguably the treaty adds little. Perhaps the greater contribution of treaties is to authorize an administrative dispute resolution mechanism that does not require court intervention. The mutual agreement procedure, authorizing the competent authorities of the two countries to resolve tax disputes relating to the treaty, is a major feature in the resolution of international disputes that threaten double taxation. While many countries do not break out their mutual agreement procedure case statistics reported to the OECD by category of case, the most frequent use of the mutual agreement procedure in Mexico and the United States is to resolve transfer pricing differences of the two countries (OECD, 2015f).

For many years, the mutual agreement procedure was not required to achieve a resolution. Increasingly mutual agreement cases under treaties are subject to mandatory arbitration if not otherwise resolved. While experience with arbitration in tax matters remains limited, preliminary indications are that it is working well in the cases arbitrated to date. It is increasingly common to use a form of arbitration in which each party suggests a resolution and the arbitrators have to choose one or the other (known in the U.S. as “baseball arbitration”; IRS, 2010). This is intended to force the countries to moderate extreme positions and limits the scope for arbitrator discretion. The mutual agreement process has become an important part of the international legal landscape in mitigating potential double taxation.

6.5.3 Concession agreement arbitration

As discussed in the preceding paragraph, some countries are starting to adopt mandatory arbitration in bilateral income tax treaties. Resource concession agreements routinely include arbitration clauses that in most cases would apply automatically in the event of a failure to agree on transfer pricing.26 Concession agreement arbitration clauses, however, routinely utilize procedures for traditional commercial arbitration. If the procedures used under treaties for transfer pricing cases prove to be efficient and fair (by comparison with litigation), countries should consider adapting similar procedures for transfer pricing cases in concession agreement arbitration clauses. Some commentators encourage developing countries to take a cautious approach to arbitration generally (Lennard, 2014).

6.6 Other treaty considerations

Treaties do little directly to reduce the threat of international double non-taxation. The exchange of information article, however, is a valuable enforcement tool. It authorizes exchange of tax information on an administrative basis, subject to the protections of domestic law, without intervention of a court process. In recent years, a tax information exchange agreement (“TIEA”) can accomplish the same objective without having the rest of the provisions of a bilateral income tax treaty.

A host country treaty with a low-tax treaty partner – or a high-tax treaty partner that allows its tax base to be eroded or compromised – can contribute significantly to tax avoidance. Thus, for example, a number of Sub-Saharan countries have treaties with Mauritius, notwithstanding that it is possible to achieve very low rates of tax in Mauritius. For example, if a Mauritius company that does not have a permanent establishment charges a host country affiliate a fee for services, Article 7 of the UN Model Treaty prevents taxation of the “business profit,” a deduction normally is allowed for the payment and the income will be taxed at a very low rate in Mauritius. (This has the same result as described in Example 4.) This creates a substantial incentive for the MNE taxpayer to overstate the management services fee that the Mauritius affiliate charges the host country affiliate. One alternative is for the host country to impose and to include a treaty provision that allows a withholding tax on services consumed in the host country. If it is necessary for host country to have a treaty with a country whose tax law or practice offers such scope for tax avoidance, consideration should be given to adopting a robust limitation of benefits provision that would limit its benefits to persons with substantial economic nexus in the treaty partner country and an anti-abuse provision to cover abusive cases not covered by this provision (OECD, 2015b).

There has been increasing question about the utility of treaties for developing countries (Dagan, 2000; Thuronyi, 2001). One proposal might be to limit a host country treaty to the same scope as a tax information exchange agreement (TIEA) plus an Article 9 arm’s length pricing commitment and an Article 26 mutual agreement procedure (Thuronyi, 2010). This would achieve most of the benefits of an income tax treaty without sacrificing the source country tax base.

6.7 Transfer pricing enforcement

It is beyond the scope of this chapter to discuss enforcement, but clearly a critical element of an overall response to transfer pricing is to build an enforcement capacity and prioritize use of resources (see Chapter 4 by Calder, this volume). The starting point of a transfer pricing enforcement strategy is an effective risk-based approach to allocating resources, training and capacity building and winning initial cases to build morale and achieve deterrence objectives. As described, for very important specialist issues, it is critical to retain an industry expert. There are useful resources for guidance and technical assistance in this area (OECD, 2013c).

7 Conclusion

Transfer pricing is a challenge for governments in resource-rich developing countries, but it is a challenge that can be addressed and revenue loss mitigated. The amounts at stake in the context of mineral extraction are material. Host countries should draw on strategies that have been shown to work and adapt them to their specific context. It is possible to address transfer pricing challenges and foster a climate for investment with smart tax system design, disciplined administration of transfer pricing rules and tough but fair enforcement.


    ActionAid. (2010) Calling Time: Why SABMiller should Stop Dodging Taxes in Africa. Available at [Accessed November 22 2015].

    ActionAid. (2013) Sweet Nothings: The Human Cost of a British Sugar Giant Avoiding Taxes in Southern Africa. Available at [Accessed November 22 2015].

    AltshulerR. and H.Grubert. (2010) “Formula Apportionment: Is it Better than the Current System and Are There Better Alternatives?National Tax Journal 63(1145) pp. 11451184.

    AndrusJ.M.Bennett and C.Silberztein. (2011) “The Arm’s Length Principle and Developing EconomiesTransfer Pricing Report (BNA) (October).

    AuerbachA.M.P.Devereux and H.Simpson. (2010) “Taxing Corporate Income” in J.MirrleesS.AdamT.BesleyR.BlundellS.BondR.ChoteM.GammieP.JohnsonG.Myles and J.Poterba (eds) Dimensions of Tax Design: The Mirrlees Review (Oxford) pp. 837893.

    Avi-YonahR.K.A.Clausing and M.C.Durst. (2009) “Allocating Business Profits for Tax Purposes: A Proposal to Adopt a Formulary Profit SplitFlorida Tax Review9 (497) 497553.

    BenshalomI. (2013) “Rethinking the Source of the Arm’s Length Transfer Pricing ProblemVirginia Tax Review32 (425) 425460.

    BerginT. (2012a) “How Starbucks Avoids UK TaxesReuters [online] Special Report Available at [Accessed November 25 2015].

    BoadwayR. and M.Keen. (2010) “Theoretical Perspectives on Resource Tax Design” in P.DanielM.Keen and C.McPherson (eds) The Taxation of Petroleum and Minerals: Principles Problems and Practice (London, New York: Routledge) pp. 1374.

    ClausingK.A. (2011) “The Revenue Effects of Multinational Firm Income ShiftingTax Notes 130(1580) 15801586.

    Columbia Center on Sustainable Investment (CCSI). (2015) ResourceContracts.Org: A Database of Publicly Available Oil Gas and Mining Contracts Available at [Accessed November 26 2015].

    DaganT. (2000) “The Treaties MythNew York University Journal of Law and Politics32 (939) 939996.

    DurstM.C. (2010) “Making Transfer Pricing Work for Developing CountriesTax Notes International60(851).

    European Commission COM. (2011) 121/3. Proposal for a Council Directive on a Common Consolidated Corporate Tax Base.

    FlemingJ.R.Peroni and S.Shay. (2014) “Formulary Apportionment in the U.S. International Income Tax System: Putting Lipstick on a Pig?Michigan Journal of International Law36 (1) 157.

    FlemingJ.R.Peroni and S.Shay. (2015) “Getting Serious About Cross-Border Earnings Stripping: Establishing an Analytical FrameworkUniversity of North Carolina Law Journal93 (763) 673742.

    FuestC. and N.Reidel. (2010) Tax Evasion and Tax Avoidance in Developing Countries: The Role of International Profit Shifting (Oxford University Centre for Bus. Tax’n Working Paper 10/12).

    GraetzM. and A.Warren. (2006) “Income Tax Discrimination and the Economic Integration of EuropeYale Law Journal115(1186) 11861255.

    GruberJ. (2011) Public Finance and Public Policy3rd edition. Worth PublishersNew York.

    GrubertH. (2005) “Tax Credits, Source Rules, Trade, and Electronic Commerce: Behavioral Margins and the Design of International Tax SystemsTax Law Review58 (149) 149190.

    GrubertH. (2012) “Foreign Taxes and the Growing Share of U.S. Multinational Company Income Abroad: Profits, Not Sales, Are Being GlobalizedNational Tax Journal65 (247) 247282.

    HellersteinW. (2012) “Tax Planning under the CCCTB’s Formulary Apportionment Provisions: The Good, the Bad and the Ugly” in D.Weber (ed) Kluwer: CCCTB Selected Issues. Page range is 221252.

    HinesJ. (2010) “Income Misattribution Under Formula ApportionmentEuropean Economic Review54 (108) 108120.

    IMF. (2011) Revenue Mobilization in Developing Countries Policy Paper (March 11 2011).

    IMF. (2012) Fiscal Regimes for Extractive Industries: Design and Implementation.

    IMF. (2014) Spillovers in International Corporate Taxation Policy Paper (May 9 2014).

    IMF OECD UN and World Bank. (2011) “Supporting the Development of More Effective Tax SystemsReport to the G-20 Development Working Group.

    IRS. (2006) IRS Accepts Settlement Offer in Largest Transfer Pricing Dispute (IR-2006–142) Available at [Accessed November 22 2015].

    IRS. (2010) U.S. – Canada Arbitration Board Operating Guidelines Available at–10_final.pdf [Accessed November 22 2015].

    IRS. (2015) Schedule UTP TY2013 Filing Statistics Available at [Accessed November 22 2015].

    IsenberghJ. (2010) International Taxation (New York: Foundation Press).

    LennardM. (2014) “Transfer Pricing Arbitration as an Option for Developing CountriesIntertax42(179).

    LodinS. (2013) “Intragroup Royalties as a Vehicle for International Tax ArbitrageTax Notes International71(1317) 13171319.

    MorseS. (2010) “Revisiting Global Formulary ApportionmentVirginia Tax Review29(594).

    Netherlands Decree of 18 December 2013 Stb. 569 2013 (December 30 2013)593644.

    NevilleS. and TreanorJ. (2012) “Starbucks to Pay £20m in Tax Over Next Two Years After Customer RevoltThe Guardian [Online] Available at [Accessed November 22 2015].

    OECD. (2010) Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 2010 (Paris: OECD Publishing). Available at:

    OECD. (2011) Transfer Pricing Legislation – a Suggested Approach. (Paris: OECD Publishing) Available at:

    OECD. (2012) Dealing Effectively with the Challenges of Transfer PricingOECD PublishingParis.

    OECD. (2013a) Addressing Base Erosion and Profit ShiftingOECD PublishingParis.

    OECD. (2013b) Revised Section E on Safe Harbors in Chapter IV of the Transfer Pricing GuidelinesOECD PublishingParis.

    OECD. (2013c) Draft Handbook on Transfer Pricing Risk AssessmentOECD PublishingParis.

    OECD. (2013d) Action Plan on Base Erosion and Profit ShiftingOECD PublishingParis.

    OECD. (2015a) Neutralising the Effects of Hybrid Mismatch Arrangements Action 2–2015 Final Report OECD/G20 Base Erosion and Profit Shifting ProjectOECD PublishingParis.

    OECD. (2015b) Preventing the Granting of Treaty Benefits in Inappropriate Circumstances Action 6–2015 Final Report OECD/G20 Base Erosion and Profit Shifting ProjectOECD PublishingParis.

    OECD. (2015c) Aligning Transfer Pricing Outcomes with Value Creation Actions 8–10 2015 Final Report OECD/G20 Base Erosion and Profit Shifting ProjectOECD PublishingParis.

    OECD. (2015d) Limiting Base Erosion Involving Interest Deductions and Other Financial Payments Action 4–2015 Final Report OECD/G20 Base Erosion and Profit Shifting ProjectOECD PublishingParis.

    OECD. (2015e) Transfer Pricing Documentation and Country-by-Country Reporting Action 13–2015 Final Report OECD/G20 Base Erosion and Profit Shifting ProjectOECD PublishingParis.

    OECD. (2015f) Mutual Agreement Procedure Statistics 2006–2014 Available at–2014.htm#US [Accessed November 27 2015].

    Platts. (2015) Methodology and Specifications Guide Available at [Accessed November 22 2015].

    RoinJ. (2008) “Can the Income Tax Be Saved – The Promise and Pitfalls of Adopting Worldwide Formulary ApportionmentTax Law Review61 (169) 169240.

    SchatanRoberto. (2012) “Tax-Minimizing Strategies and the Arm’s Length PrincipleTax Notes International65(121) 121126.

    ShayS. (2009) “Ownership Neutrality and Practical ComplicationsTax Law Review62 (317) 317332.

    SlemrodJ. (2010) “Location, (Real) Location, (Tax) Location: An Essay on Mobility’s Place in Optimal TaxationNational Tax Journal63 (843) 843864.

    Tanzania Income Tax Act (Rev. Ed. 2006).

    ThuronyiV. (2001) “International Tax Cooperation and a Multilateral TreatyBrooklyn Journal of International Law26(1641) 16411681.

    ThuronyiV. (2010) “Tax Treaties and Developing Countries” in M.LangP.PistoneJ.SchuchC.StaringerA.Storck & M.Zagler (eds) Tax Treaties: Building Bridges between Law and Economics (IBFD: Online books) n.p.

    TuerffT.SierraG.TrumpC. and NarayanM. (2011) Overview of Outbound Tax Planning for U.S. Multinational Corporations Tax Management International 40 (No. 7).

    Uganda. (2011) The Income Tax (Transfer Pricing) Regulations. Author.

    UK House of Commons. (2012) Committee of Public Accounts HM Revenue & Customs: Annual Report and Accounts 2011–12 Nineteenth Report of Session 2012–13 (HC 716) Available at [Accessed November 22 2015].

    U.S. Senate. (2012) Permanent Subcommittee on Investigations of the Committee on Homeland Security and Government Affairs Hearing on Offshore Profit Shifting and the U.S. Tax Code (Part 1 Hewlett Packard and Microsoft) Memorandum from Chairman Carl Levin and Senator Tom Coburn to Subcommittee Members Offshore Profit Shifting and the Internal Revenue Code. Available at [Accessed May 30 2013].

    U.S. Senate. (2013) Permanent Subcommittee on Investigations of the Committee on Homeland Security and Government Affairs Hearing on Offshore Profit Shifting and the U.S. Tax Code – Part 2 (Apple Inc.) Memorandum from Chairman Carl Levin and Senator John McCain to Subcommittee Members Offshore Profit Shifting and the Internal Revenue Code – Part 2 (Apple Inc.) Available at [Accessed May 30 2013].

    UN. (2000) (Millennium Declaration) Resolution 55/2 (September 18 2000).

    UN. (2002) International Conference on Financing for Development, (A/CONF. 198/11). UN. (2013)Practical Manual on Transfer Pricing for Developing Countries.

    WTO. (2010) World Trade Report.

    WTO. (2014) World Trade Report.


Mr. Shay thanks Joseph Bell, Jack Calder, Joseph Guttentag, Michael Keen, Leandra Lederman, Michael Lennard, Emil Sunley, Artur Świstak and participants at a Harvard Law School Workshop on Current Research in Taxation for comments on earlier drafts. He also thanks the Harvard Law School and the Harvard Fund for Tax and Fiscal Policy Research for summer research support. Mr. Shay discloses certain activities not connected with his position at Harvard Law School, one or more of which may relate to the subject matter of this article, at = 10794. The views expressed in this article are those of Mr. Shay and do not reflect those of Harvard University, any client of Mr. Shay, or any organization for which he serves as an officer or renders pro bono services.

In some intermediary cases, a third country may impose tax under a controlled foreign company (CFC) regime on the income of the intermediary company and thereby also have a tax interest in the transfer pricing. In practice, this issue rarely arises largely because of the ability to plan around or otherwise avoid CFC regimes. If, as suggested in the OECD’s BEPS action plan, CFC regimes are strengthened, provision should be made (in treaties or other agreements) for the host country and the ultimate residence country to reach a mutual agreement on how each country is pricing transactions with the intermediary company to avoid double taxation or double non-taxation (OECD, 2013d).

Joel Slemrod refers to the ability to use real investment to credibly support profit shifting as an “avoidance-facilitating effect” of real decisions (Slemrod, 2010, p. 856).

If the effective tax in a host country is reduced by tax avoidance planning, then investments with a lower pre-tax return may be made that would not otherwise be made. This result disadvantages the alternative investment with a higher pre-tax return – the alternative investment may not be made if the multinational can earn a higher after-tax return by using non–arm’s length transfer pricing. Foregoing the alternative investment may be favorable for the host country if the alternative investment is not in the host country and depending on the other consequences of the unintended erosion in tax base. The non–arm’s length outcome may be unfavorable for the host country if the revenue loss over the life of the project does not justify the marginal gain from the project.

A response by some countries to objections to treaty shopping and other practices using third-country intermediary structures is to require more “people” functions in the country where the intermediary entity is located (Netherlands, 2013). This is perceived to be in response to pressure from the OECD project on base erosion and profit shifting (BEPS; OECD, 2013a).

One could argue as well that actions by taxpayers such as making voluntary tax payments is an ineffective or counterproductive response at least from the perspective of tax policy. Tax laws should not operate based on the marketing objectives of businesses or on the impulses, however well meaning, of taxpayers to pay a larger share of the costs of public goods (Neville and Treanor, 2012).

As Professor Isenbergh notes, “An arm’s length price has in common with a policeman that as often as not you can’t find one when you need one” (Isenbergh, 2010, p. 68).

It is possible that formulas could be adjusted, for example to include discovery value of a concession in the property factor, for extractive industries. This would exacerbate the difficulties of implementation. A number of commentators argue that formula apportionment would not improve on arm’s length separate accounting (Andrus, Bennett and Silberztein, 2011; Altshuler and Grubert, 2010; Hines, 2010; Morse, 2010; Roin, 2008). Recently prominent commentators favor a sales-only formula (Avi-Yonah, Clausing and Durst, 2009).

Durst recommends that countries use safe harbor profit ranges for these activities in order to simplify administration and conserve resources. The OECD recently modified the Transfer Pricing Guidelines’ discussion of safe harbors to be open to the possibility that in certain circumstances they may be warranted (OECD, 2013b). If not properly limited, there is a risk that a safe harbor becomes either a floor or a ceiling and operates against the interest of the government.

Where the sale to an affiliate is immediately followed by an uncontrolled sale to an unrelated purchaser, it generally is possible to make reference to the uncontrolled sale price. We do not discuss here the problem of accommodation or “straw” uncontrolled buyers who in turn resell to affiliates of the taxpayer or have offsetting purchase arrangements. Such intermediate or arranged sales of course should not be considered uncontrolled sales.

In mining, royalty rates generally are a relatively small fraction of typical income tax rates but normally are applied to gross sales at the time of production. Even taking into account the time value of money effect of income tax collections coming later in the life of a project than royalty receipts, in many cases the income tax is expected to yield more revenue than the royalty.

Some countries use a production sharing regime instead of a royalty; however, the pricing is equally important.

In cases where there is a market for hedging price risk of the mineral in question, there is a further issue of how to address gains and losses. There have been examples that suggest mining producers are consistently on the loss side of hedging transactions raising the concern that there may be offsetting positions held by offshore affiliates. For this reason, some countries “ring-fence” hedging gains and losses so that they do not affect the royalty or income tax base – see Chapter 4 by Calder, this volume, at p. 83.

CFR means cost and freight to the named port of destination. The seller pays the costs of moving the goods to destination, however, risk transfers to the buyer once the goods are placed on board the ship for transport so the buyer bears risks of loss or damage. The term generally applies to maritime transport.

FOB stands for “free on board.” The seller delivers the goods on board the ship. From that point, the buyer bears all costs and risks of loss or damage.

A marketing allowance raises a number of issues, including whether it is appropriate; if appropriate, whether it properly is in the sales price or would be separately compensated; and if separately compensated, whether the amount or what amount properly is allowable as a host country deduction. If management fees are charged to the host country affiliate, the fee may take account of the costs of the business’s marketing organization. As a general principle, a host country should resist adjustments that are not based on transparent information or third-party information.

There may be fewer issues for oil than for minerals. The differences in quality of different crude oils may be more readily accepted by or adapted to by producers and the quality of oil from a reservoir generally is consistent. There are quoted prices for “benchmark” crudes, including liquid forward markets, and standard methods for adjusting for quality differences. The costs between the wellhead and the delivery terminal where oil is most often sold at arm’s length are limited and measurable – see Chapter 4 by Calder, this volume, at pp. 90–92.

For a major extractive industry MNE, the costs of forming and using entities are spread over multiple projects. It often is possible to use existing entities for “double purpose” as a conduit for income shifting while relying on “substance” of other business activity to support recognition of the intermediary entity.

The interest which is paid by issuing additional debt often is referred to as “payment in kind” or PIK debt. PIK debt can preserve cash while maximizing the tax deduction for interest.

At first blush, this would seem to be at variance with the position stated in the UN Practical Manual on Transfer Pricing, which states, “For developing countries, transfer pricing rules are essential to provide a climate of certainty and an environment for increased cross-border trade while at the same time ensuring that the country is not losing out on critical tax revenue. Transfer pricing is of paramount importance and hence detailed transfer pricing rules are essential” (UN, 2013, ¶1.2.1 3). As discussed in the text, it simply is not correct that detailed transfer pricing rules are essential. Potentially affected taxpayers who are seeking certainty and tax authorities who find courts reluctant to support broad discretionary authority may find them useful.

The UN Manual’s discussion of penalties is confined largely to the use of penalties as an incentive to provide transfer pricing documentation and no discussion of penalties as a disincentive to partake of aggressive transfer pricing. It is unclear why it did not consider penalty regimes that would provide a disincentive to aggressive pricing, though it is possible that it was assumed that tax understatement penalties would apply automatically. This is unlikely if fraud or negligence is a condition for the penalty.

The determination whether a penalty threshold is exceeded does not take account of an adjustment if contemporaneous documentation establishes that the method used was reasonable. It is interesting to consider the range of cases in which it would be true that the pricing should be subject to an adjustment but the incorrect price used by the taxpayer was nonetheless reasonable and therefore should not be subject to a penalty. It seems unlikely that the tax authority would acknowledge that the documentation was adequate in a case in which it also makes an adjustment.

The UN Manual includes a chapter on documentation (UN, 2013, Ch. 7). Transfer pricing documentation generally should include information about (i) the associated enterprises engaged in controlled transactions, (ii) the nature and terms of the controlled transactions, (iii) business structure and operating results of the enterprises, (iv) potentially comparable uncontrolled transactions, (v) pricing, business strategies and special circumstances relevant to the controlled transactions (including set-offs) and (vi) other information relevant to evaluating the functions performed, assets employed and risks taken by the controlled parties to the transactions (UN, 2013, ¶¶–5).

In the context of extractive industry concessions, it is important that the concession agreements do not foreclose the ability to make adjustments and assert penalties.

This chapter does not consider the broader issue of whether stabilization clauses should apply to taxes in the first place. Among other considerations, applying stabilization to taxes places substantial burdens on tax authorities in having to enforce different generations of tax laws depending on when a concession was entered into.

It would be a matter of domestic law whether an applicable treaty mutual agreement procedure would take precedence over concession’s arbitration clause and be finally binding. The parties clearly could contractually agree to be bound by a treaty procedure.

    Other Resources Citing This Publication