Category: Chapter IV: Administrative Dispute Resolution
This chapter examines various administrative procedures that could be applied to minimise transfer pricing disputes and to help resolve them when they do arise between taxpayers and their tax administrations, and between different tax administrations. Such disputes may arise even though the guidance in these Guidelines is followed in a conscientious effort to apply the arm’s length principle. It is possible that taxpayers and tax administrations may reach differing determinations of the arm’s length conditions for the controlled transactions under examination given the complexity of some transfer pricing issues and the difficulties in interpreting and evaluating the circumstances of individual cases.
Where two or more tax administrations take different positions in determining arm’s length conditions, double taxation may occur. Double taxation means the inclusion of the same income in the tax base by more than one tax administration, when either the income is in the hands of different taxpayers (economic double taxation, for associated enterprises) or the income is in the hands of the same juridical entity (juridical double taxation, for permanent establishments). Double taxation is undesirable and should be eliminated whenever possible, because it constitutes a potential barrier to the development of international trade and investment flows. The double inclusion of income in the tax base of more than one jurisdiction does not always mean that the income will actually be taxed twice.
This chapter discusses several administrative approaches to resolving disputes caused by transfer pricing adjustments and for avoiding double taxation. Section B discusses transfer pricing compliance practices by tax administrations, in particular examination practices, the burden of proof, and penalties. Section C discusses corresponding adjustments (Paragraph 2 of Article 9 of the OECD Model Tax Convention) and the mutual agreement procedure (Article 25). Section D describes the use of simultaneous tax examinations by two (or more) tax administrations to expedite the identification, processing, and resolution of transfer pricing issues (and other international tax issues). Sections E and F describe some possibilities for minimising transfer pricing disputes between taxpayers and their tax administrations. Section E addresses the possibility of developing safe harbours for certain taxpayers, and Section F deals with advance pricing arrangements, which address the possibility of determining in advance a transfer pricing methodology or conditions for the taxpayer to apply to specified controlled transactions. Section G considers briefly the use of arbitration procedures to resolve transfer pricing disputes between countries.
Tax compliance practices are developed and implemented in each member country according to its own domestic legislation and administrative procedures. Many domestic tax compliance practices have three main elements: a) to reduce opportunities for non-compliance (e.g. through withholding taxes and information reporting); b) to provide positive assistance for compliance (e.g. through education and published guidance); and, c) to provide disincentives for non-compliance. As a matter of domestic sovereignty and to accommodate the particularities of widely varying tax systems, tax compliance practices remain within the province of each country. Nevertheless a fair application of the arm’s length principle requires clear procedural rules to ensure adequate protection of the taxpayer and to make sure that tax revenue is not shifted to countries with overly harsh procedural rules. However, when a taxpayer under examination in one country is a member of an MNE group, it is possible that the domestic tax compliance practices in a country examining a taxpayer will have consequences in other tax jurisdictions. This may be particularly the case when cross-border transfer pricing issues are involved, because the transfer pricing has implications for the tax collected in the tax jurisdictions of the associated enterprises involved in the controlled transaction. If the same transfer pricing is not accepted in the other tax jurisdictions, the MNE group may be subject to double taxation as explained in paragraph 4.2. Thus, tax administrations should be conscious of the arm’s length principle when applying their domestic compliance practices and the potential implications of their transfer pricing compliance rules for other tax jurisdictions, and seek to facilitate both the equitable allocation of taxes between jurisdictions and the prevention of double taxation for taxpayers.
This section describes three aspects of transfer pricing compliance that should receive special consideration to help tax jurisdictions administer their transfer pricing rules in a manner that is fair to taxpayers and other jurisdictions. While other tax law compliance practices are in common use in OECD member countries – for example, the use of litigation and evidentiary sanctions where information may be sought by a tax administration but is not provided – these three aspects will often impact on how tax administrations in other jurisdictions approach the mutual agreement procedure process and determine their administrative response to ensuring compliance with their own transfer pricing rules. The three aspects are: examination practices, the burden of proof, and penalty systems. The evaluation of these three aspects will necessarily differ depending on the characteristics of the tax system involved, and so it is not possible to describe a uniform set of principles or issues that will be relevant in all cases. Instead, this section seeks to provide general guidance on the types of problems that may arise and reasonable approaches for achieving a balance of the interests of the taxpayers and tax administrations involved in a transfer pricing inquiry.
Examination practices vary widely among OECD member countries. Differences in procedures may be prompted by such factors as the system and the structure of the tax administration, the geographic size and population of the country, the level of domestic and international trade, and cultural and historical influences.
Transfer pricing cases can present special challenges to the normal audit or examination practices, both for the tax administration and for the taxpayer. Transfer pricing cases are fact-intensive and may involve difficult evaluations of comparability, markets, and financial or other industry information. Consequently, a number of tax administrations have examiners who specialise in transfer pricing, and transfer pricing examinations themselves may take longer than other examinations and follow separate procedures.
Because transfer pricing is not an exact science, it will not always be possible to determine the single correct arm’s length price; rather, as Chapter III recognises, the correct price may have to be estimated within a range of acceptable figures. Also, the choice of methodology for establishing arm’s length transfer pricing will not often be unambiguously clear. Taxpayers may experience particular difficulties when the tax administration proposes to use a methodology, for example a transactional profit method, that is not the same as that used by the taxpayer.
In a difficult transfer pricing case, because of the complexity of the facts to be evaluated, even the best-intentioned taxpayer can make an honest mistake. Moreover, even the best-intentioned tax examiner may draw the wrong conclusion from the facts. Tax administrations are encouraged to take this observation into account in conducting their transfer pricing examinations. This involves two implications. First, tax examiners are encouraged to be flexible in their approach and not demand from taxpayers in their transfer pricing a precision that is unrealistic under all the facts and circumstances. Second, tax examiners are encouraged to take into account the taxpayer’s commercial judgment about the application of the arm’s length principle, so that the transfer pricing analysis is tied to business realities. Therefore, tax examiners should undertake to begin their analyses of transfer pricing from the perspective of the method that the taxpayer has chosen in setting its prices. The guidance provided in Chapter II, Part I dealing with the selection of the most appropriate transfer pricing method also may assist in this regard.
A tax administration should keep in mind in allocating its audit resources the taxpayer’s process of setting prices, for example whether the MNE group operates on a profit centre basis. See paragraph 1.5.
Like examination practices, the burden of proof rules for tax cases also differ among OECD member countries. In most jurisdictions, the tax administration bears the burden of proof both in its own internal dealings with the taxpayer (e.g. assessment and appeals) and in litigation. In some of these countries, the burden of proof can be reversed, allowing the tax administration to estimate taxable income, if the taxpayer is found not to have acted in good faith, for example, by not cooperating or complying with reasonable documentation requests or by filing false or misleading returns. In other countries, the burden of proof is on the taxpayer. In this respect, however, the conclusions of paragraphs 4.16 and 4.17 should be noted.
The implication for the behaviour of the tax administration and the taxpayer of the rules governing burden of proof should be taken into account. For example, where as a matter of domestic law the burden of proof is on the tax administration, the taxpayer may not have any legal obligation to prove the correctness of its transfer pricing unless the tax administration makes a prima facie showing that the pricing is inconsistent with the arm’s length principle. Even in such a case, of course, the tax administration might still reasonably oblige the taxpayer to produce its records that would enable the tax administration to undertake its examination. In some countries, taxpayers have a duty to cooperate with the tax administration imposed on them by law. In the event that a taxpayer fails to cooperate, the tax administration may be given the authority to estimate the taxpayer’s income and to assume relevant facts based on experience. In these cases, tax administrations should not seek to impose such a high level of cooperation that would make it too difficult for reasonable taxpayers to comply.
In jurisdictions where the burden of proof is on the taxpayer, tax administrations are generally not at liberty to raise assessments against taxpayers which are not soundly based in law. A tax administration in an OECD member country that applies the arm’s length principle, for example, could not raise an assessment based on a taxable income calculated as a fixed percentage of turnover and simply ignore the arm’s length principle. In the context of litigation in countries where the burden of proof is on the taxpayer, the burden of proof is often seen as a shifting burden. Where the taxpayer presents to a court a reasonable argument and evidence to suggest that its transfer pricing was arm’s length, the burden of proof may legally or de facto shift to the tax administration to counter the taxpayer’s position and to present argument and evidence as to why the taxpayer’s transfer pricing was not arm’s length and why the assessment is correct. On the other hand, where a taxpayer makes little effort to show that its transfer pricing was arm’s length, the burden imposed on the taxpayer would not be satisfied where a tax administration raised an assessment which was soundly based in law.
When transfer pricing issues are present, the divergent rules on burden of proof among OECD member countries will present serious problems if the strict legal rights implied by those rules are used as a guide for appropriate behaviour. For example, consider the case where the controlled transaction under examination involves one jurisdiction in which the burden of proof is on the taxpayer and a second jurisdiction in which the burden of proof is on the tax administration. If the burden of proof is guiding behaviour, the tax administration in the first jurisdiction might make an unsubstantiated assertion about the transfer pricing, which the taxpayer might accept, and the tax administration in the second jurisdiction would have the burden of disproving the pricing. It could be that neither the taxpayer in the second jurisdiction nor the tax administration in the first jurisdiction would be making efforts to establish an acceptable arm’s length price. This type of behaviour would set the stage for significant conflict as well as double taxation.
Consider the same facts as in the example in the preceding paragraph. If the burden of proof is again guiding behaviour, a taxpayer in the first jurisdiction being a subsidiary of a taxpayer in the second jurisdiction (notwithstanding the burden of proof and these Guidelines), may be unable or unwilling to show that its transfer prices are arm’s length. The tax administration in the first jurisdiction after examination makes an adjustment in good faith based on the information available to it. The parent company in the second jurisdiction is not obliged to provide to its tax administration any information to show that the transfer pricing was arm’s length as the burden of proof rests with the tax administration. This will make it difficult for the two tax administrations to reach agreement in competent authority proceedings.
In practice, neither countries nor taxpayers should misuse the burden of proof in the manner described above. Because of the difficulties with transfer pricing analyses, it would be appropriate for both taxpayers and tax administrations to take special care and to use restraint in relying on the burden of proof in the course of the examination of a transfer pricing case. More particularly, as a matter of good practice, the burden of proof should not be misused by tax administrations or taxpayers as a justification for making groundless or unverifiable assertions about transfer pricing. A tax administration should be prepared to make a good faith showing that its determination of transfer pricing is consistent with the arm’s length principle even where the burden of proof is on the taxpayer, and taxpayers similarly should be prepared to make a good faith showing that their transfer pricing is consistent with the arm’s length principle regardless of where the burden of proof lies.
The Commentary on paragraph 2 of Article 9 of the OECD Model Tax Convention makes clear that the State from which a corresponding adjustment is requested should comply with the request only if that State “considers that the figure of adjusted profits correctly reflects what the profits would have been if the transactions had been at arm’s length”. This means that in competent authority proceedings the State that has proposed the primary adjustment bears the burden of demonstrating to the other State that the adjustment “is justified both in principle and as regards the amount.” Both competent authorities are expected to take a cooperative approach in resolving mutual agreement cases.
Penalties are most often directed toward providing disincentives for non-compliance, where the compliance at issue may relate to procedural requirements such as providing necessary information or filing returns, or to the substantive determination of tax liability. Penalties are generally designed to make tax underpayments and other types of non-compliance more costly than compliance. The Committee on Fiscal Affairs has recognised that promoting compliance should be the primary objective of civil tax penalties. OECD Report Taxpayers’ Rights and Obligations (1990). If a mutual agreement results in a withdrawal or reduction of an adjustment, it is important that there exist possibilities to cancel or mitigate a penalty imposed by the tax administrations.
Care should be taken in comparing different national penalty practices and policies with one another. First, any comparison needs to take into account that there may be different names used in the various countries for penalties that accomplish the same purposes. Second, the overall compliance measures of an OECD member country should be taken into account. National tax compliance practices depend, as indicated above, on the overall tax system in the country, and they are designed on the basis of domestic need and balance, such as the choice between the use of taxation measures that remove or limit opportunities for noncompliance (e.g. imposing a duty on taxpayers to cooperate with the tax administration or reversing the burden of proof in situations where a taxpayer is found not to have acted in good faith) and the use of monetary deterrents (e.g. additional tax imposed as a consequence of underpayments of tax in addition to the amount of the underpayment). The nature of tax penalties may also be affected by the judicial system of a country. Most countries do not apply no-fault penalties; in some countries, for example, the imposition of a no-fault penalty would be against the underlying principles of their legal system.
There are a number of different types of penalties that tax jurisdictions have adopted. Penalties can involve either civil or criminal sanctions – criminal penalties are virtually always reserved for cases of very significant fraud, and they usually carry a very high burden of proof for the party asserting the penalty (i.e. the tax administration). Criminal penalties are not the principal means to promote compliance in any of the OECD member countries. Civil (or administrative) penalties are more common, and they typically involve a monetary sanction (although as discussed above there may be a non-monetary sanction such as a shifting of the burden of proof when, e.g. procedural requirements are not met or the taxpayer is uncooperative and an effective penalty results from a discretionary adjustment).
Some civil penalties are directed towards procedural compliance, such as timely filing of returns and information reporting. The amount of such penalties is often small and based on a fixed amount that may be assessed for each day in which, e.g. the failure to file continues. The more significant civil penalties are those directed at the understatement of tax liability.
Although some countries may refer to a “penalty”, the same or similar imposition by another country may be classified as “interest”. Some countries’ “penalty” regimes may therefore include an “additional tax”, or “interest”, for understatements which result in late payments of tax beyond the due date. This is often designed to ensure the revenue recovers at least the real time value of money (taxes) lost.
Civil monetary penalties for tax understatement are frequently triggered by one or more of the following: an understatement of tax liability exceeding a threshold amount, negligence of the taxpayer, or wilful intent to evade tax (and also fraud, although fraud can trigger much more serious criminal penalties). Many OECD member countries impose civil monetary penalties for negligence or willful intent, while only a few countries penalise “no-fault” understatements of tax liability.
It is difficult to evaluate in the abstract whether the amount of a civil monetary penalty is excessive. Among OECD member countries, civil monetary penalties for tax understatement are frequently calculated as a percentage of the tax understatement, where the percentage most often ranges from 10% to 200%. In most OECD member countries, the rate of the penalty increases as the conditions for imposing the penalty increase. For instance, the higher rate penalties often can be imposed only by showing a high degree of taxpayer culpability, such as a wilful intent to evade. “No-fault” penalties, where used, tend to be at lower rates than those triggered by taxpayer culpability (see paragraph 4.28).
Improved compliance in the transfer pricing area is of some concern to OECD member countries and the appropriate use of penalties may play a role in addressing this concern. However, owing to the nature of transfer pricing problems, care should be taken to ensure that the administration of a penalty system as applied in such cases is fair and not unduly onerous for taxpayers.
Because cross-border transfer pricing issues implicate the tax base of two jurisdictions, an overly harsh penalty system in one jurisdiction may give taxpayers an incentive to overstate taxable income in that jurisdiction contrary to Article 9. If this happens, the penalty system fails in its primary objective to promote compliance and instead leads to non-compliance of a different sort – non-compliance with the arm’s length principle and under-reporting in the other jurisdiction. Each OECD member country should ensure that its transfer pricing compliance practices are not enforced in a manner inconsistent with the objectives of the OECD Model Tax Convention, avoiding the distortions noted above.
It is generally regarded by OECD member countries that the fairness of the penalty system should be considered by reference to whether the penalties are proportionate to the offence. This would mean, for example, that the severity of a penalty would be balanced against the conditions under which it would be imposed, and that the harsher the penalty the more limited the conditions in which it would apply.
Since penalties are only one of many administrative and procedural aspects of a tax system, it is difficult to conclude whether a particular penalty is fair or not without considering the other aspects of the tax system. Nonetheless, OECD member countries agree that the following conclusions can be drawn regardless of the other aspects of the tax system in place in a particular country. First, imposition of a sizable “no-fault” penalty based on the mere existence of an understatement of a certain amount would be unduly harsh when it is attributable to good faith error rather than negligence or an actual intent to avoid tax. Second, it would be unfair to impose sizable penalties on taxpayers that made a reasonable effort in good faith to set the terms of their transactions with associated enterprises in a manner consistent with the arm’s length principle. In particular, it would be inappropriate to impose a transfer pricing penalty on a taxpayer for failing to consider data to which it did not have access, or for failure to apply a transfer pricing method that would have required data that was not available to the taxpayer. Tax administrations are encouraged to take these observations into account in the implementation of their penalty provisions.
The mutual agreement procedure is a well-established means through which tax administrations consult to resolve disputes regarding the application of double tax conventions. This procedure, described and authorised by Article 25 of the OECD Model Tax Convention, can be used to eliminate double taxation that could arise from a transfer pricing adjustment.
Article 25 sets out three different areas where mutual agreement procedures are generally used. The first area includes instances of “taxation not in accordance with the provisions of the Convention” and is covered in paragraphs 1 and 2 of the Article. Procedures in this area are typically initiated by the taxpayer. The other two areas, which do not necessarily involve the taxpayer, are dealt with in paragraph 3 and involve questions of “interpretation or application of the Convention” and the elimination of double taxation in cases not otherwise provided for in the Convention. Paragraph 10 of the Commentary on Article 25 makes clear that Article 25 is intended to be used by competent authorities in resolving not only problems of juridical double taxation but also those of economic double taxation arising from transfer pricing adjustments made pursuant to paragraph 1 of Article 9.
Paragraph 5 of Article 25, which was incorporated in the OECD Model Tax Convention in 2008, provides that, in mutual agreement procedure cases in which the competent authorities are unable to reach an agreement within two years of the initiation of the case under paragraph 1 of Article 25, the unresolved issues will, at the request of the person who presented the case, be resolved through an arbitration process. This extension of the mutual agreement procedure ensures that where the competent authorities cannot reach an agreement on one or more issues that prevent the resolution of a case, a resolution of the case will still be possible by submitting those issues to arbitration. Where one or more issues have been submitted to arbitration in accordance with such a provision, and unless a person directly affected by the case does not accept the mutual agreement that implements the arbitration decision, that decision shall be binding on both States, the taxation of any person directly affected by the case will have to conform with the decision reached on the issues submitted to arbitration and the decisions reached in the arbitral process will be reflected in the mutual agreement that will be presented to these persons. Where a particular bilateral treaty does not contain an arbitration provision similar to paragraph 5 of Article 25, the competent authorities are not obliged to reach an agreement to resolve their dispute; paragraph 2 of Article 25 requires only that the competent authorities “endeavour … to resolve the case by mutual agreement”. The competent authorities may be unable to come to an agreement because of conflicting domestic laws or restrictions imposed by domestic law on the tax administration’s power of compromise. Even in the absence of a mandatory binding arbitration provision similar to paragraph 5 of Article 25 in a particular bilateral treaty, however, the competent authorities of the Contracting States may by mutual agreement establish a binding arbitration procedure for general application or to deal with a specific case (see paragraph 69 of the Commentary on Article 25 of the OECD Model Tax Convention). It should also be noted that a multilateral Convention on the elimination of double taxation in connection with the adjustment of profits of associated enterprises2 (the Arbitration Convention) was signed by the Member States of the European Communities on 23 July 1990; the Arbitration Convention, which entered into force on 1 January 1995, provides for an arbitration mechanism to resolve transfer pricing disputes between European Union Member States.
To eliminate double taxation in transfer pricing cases, tax administrations may consider requests for corresponding adjustments as described in paragraph 2 of Article 9. A corresponding adjustment, which in practice may be undertaken as part of the mutual agreement procedure, can mitigate or eliminate double taxation in cases where one tax administration increases a company’s taxable profits (i.e. makes a primary adjustment) as a result of applying the arm’s length principle to transactions involving an associated enterprise in a second tax jurisdiction. The corresponding adjustment in such a case is a downward adjustment to the tax liability of that associated enterprise, made by the tax administration of the second jurisdiction, so that the allocation of profits between the two jurisdictions is consistent with the primary adjustment and no double taxation occurs. It is also possible that the first tax administration will agree to decrease (or eliminate) the primary adjustment as part of the consultative process with the second tax administration, in which case the corresponding adjustment would be smaller (or perhaps unnecessary). It should be noted that a corresponding adjustment is not intended to provide a benefit to the MNE group greater than would have been the case if the controlled transactions had been undertaken at arm’s length conditions in the first instance.
Paragraph 2 of Article 9 specifically provides that the competent authorities shall consult each other if necessary to determine appropriate corresponding adjustments. This confirms that the mutual agreement procedure of Article 25 may be used to consider corresponding adjustment requests. See also paragraph 10 of the Commentary on Article 25 of the OECD Model Tax Convention (“… the corresponding adjustments to be made in pursuance of paragraph 2 of [Article 9] … fall within the scope of mutual agreement procedure, both as concerns assessing whether they are well-founded and for determining their amount.” However, the overlap between the two Articles has caused OECD member countries to consider whether the mutual agreement procedure can be used to achieve corresponding adjustments where the bilateral income tax convention between two Contracting States does not include a provision comparable to paragraph 2 of Article 9. Paragraphs 11 and 12 of the Commentary on Article 25 of the OECD Model Tax Convention expressly state the view of most OECD member countries that the mutual agreement procedure is considered to apply to transfer pricing adjustment cases, including issues of whether a corresponding adjustment should be provided, even in the absence of a provision comparable to paragraph 2 of Article 9. Paragraph 12 notes that those States that do not agree with this view in practice find means of remedying economic double taxation in most cases involving bona fide companies by making use of provisions in their domestic laws.
Under paragraph 2 of Article 9, a corresponding adjustment may be made by a Contracting State either by recalculating the profits subject to tax for the associated enterprise in that country using the relevant revised price or by letting the calculation stand and giving the associated enterprise relief against its own tax paid in that State for the additional tax charged to the associated enterprise by the adjusting State as a consequence of the revised transfer price. The former method is by far the more common among OECD member countries.
In the absence of an arbitration decision arrived at pursuant to an arbitration procedure comparable to that provided for under paragraph 5 of Article 25 which provides for a corresponding adjustment, corresponding adjustments are not mandatory, mirroring the rule that tax administrations are not obliged to reach agreement under the mutual agreement procedure. Under paragraph 2 of Article 9, a tax administration should make a corresponding adjustment only insofar as it considers the primary adjustment to be justified both in principle and in amount. The non- mandatory nature of corresponding adjustments is necessary so that one tax administration is not forced to accept the consequences of an arbitrary or capricious adjustment by another State. It also is important to maintaining the fiscal sovereignty of each OECD member country.
Once a tax administration has agreed to make a corresponding adjustment it is necessary to establish whether the adjustment is to be attributed to the year in which the controlled transactions giving rise to the adjustment took place or to an alternative year, such as the year in which the primary adjustment is determined. This issue also often raises the question of a taxpayer’s entitlement to interest on the overpayment of tax in the jurisdiction which has agreed to make the corresponding adjustment (discussed in paragraphs 4.65-4.67). The first approach is more appropriate because it achieves a matching of income and expenses and better reflects the economic situation as it would have been if the controlled transactions had been at arm’s length. However, in cases involving lengthy delays between the year covered by the adjustment and the year of its acceptance by the taxpayer or a final court decision, the tax administration should have the flexibility to agree to make corresponding adjustments for the year of acceptance of or decision on the primary adjustment. This approach would need to rely on domestic law for implementation. While not ordinarily preferred, it could be appropriate as an equitable measure in exceptional cases to facilitate implementation and to avoid time limit barriers.
Corresponding adjustments can be a very effective means of obtaining relief from double taxation resulting from transfer pricing adjustments. OECD member countries generally strive in good faith to reach agreement whenever the mutual agreement procedure is invoked. Through the mutual agreement procedure, tax administrations can address issues in a non-adversarial proceeding, often achieving a negotiated settlement in the interests of all parties. It also allows tax administrations to take into account other taxing rights issues, such as withholding taxes.
At least one OECD member country has a procedure that may reduce the need for primary adjustments by allowing the taxpayer to report a transfer price for tax purposes that is, in the taxpayer’s opinion, an arm’s length price for a controlled transaction, even though this price differs from the amount actually charged between the associated enterprises. This adjustment, sometimes known as a “compensating adjustment”, would be made before the tax return is filed. Compensating adjustments may facilitate the reporting of taxable income by taxpayers in accordance with the arm’s length principle, recognising that information about comparable uncontrolled transactions may not be available at the time associated enterprises establish the prices for their controlled transactions. Thus, for the purpose of lodging a correct tax return, a taxpayer would be permitted to make a compensating adjustment that would record the difference between the arm’s length price and the actual price recorded in its books and records.
However, compensating adjustments are not recognised by most OECD member countries, on the grounds that the tax return should reflect the actual transactions. If compensating adjustments are permitted (or required) in the country of one associated enterprise but not permitted in the country of the other associated enterprise, double taxation may result because corresponding adjustment relief may not be available if no primary adjustment is made. The mutual agreement procedure is available to resolve difficulties presented by compensating adjustments, and competent authorities are encouraged to use their best efforts to resolve any double taxation which may arise from different country approaches to such year- end adjustments.
While corresponding adjustment and mutual agreement procedures have proved to be able to resolve most transfer pricing conflicts, serious concerns have been expressed by taxpayers. For example, because transfer pricing issues are so complex, taxpayers have expressed concerns that there may not be sufficient safeguards in the procedures against double taxation. These concerns are mainly addressed with the introduction in the 2008 update of the OECD Model Tax Convention of a new paragraph 5 to Article 25 which introduces a mechanism that allows taxpayers to request arbitration of unresolved issues that have prevented competent authorities from reaching a mutual agreement within two years. There is also in the Commentary on Article 25 a favourable discussion of the use of supplementary dispute resolution mechanisms in addition to arbitration, including mediation and the referral of factual disputes to third party experts.
Taxpayers have also expressed fears that their cases may be settled not on their individual merits but by reference to a balance of the results in other cases. An established good practice is that, in the resolution of mutual agreement cases, a competent authority should engage in discussions with other competent authorities in a principled, fair, and objective manner, with each case being decided on its own merits and not by reference to any balance of results in other cases. To the extent applicable, these Guidelines and proposals detailed in the Report on BEPS Action 14 (bearing in mind the difference between the minimum standard and best practices) are an appropriate basis for the development of a principled approach. Similarly, there may be a fear of retaliation or offsetting adjustments by the country from which the corresponding adjustment has been requested. It is not the intention of tax administrations to take retaliatory action; the fears of taxpayers may be a result of inadequate communication of this fact. Tax administrations should take steps to assure taxpayers that they need not fear retaliatory action and that, consistent with the arm’s length principle, each case is resolved on its own merits. Taxpayers should not be deterred from initiating mutual agreement procedures where Article 25 is applicable.
Concerns that have been expressed regarding the mutual agreement procedure, as it affects corresponding adjustments, include the following, which are discussed separately in the sections below: Taxpayers may be denied access to the mutual agreement procedure in transfer pricing cases; Time limits under domestic law for the amendments of tax assessments may make corresponding adjustments unavailable if the relevant tax treaty does not override those limits; Mutual agreement procedure cases may take a long time; Taxpayer participation may be limited; Published guidance may not be readily available to instruct taxpayers on how the mutual agreement procedure may be used; and There may be no procedures to suspend the collection of tax deficiencies or the accrual of interest pending resolution of the mutual agreement procedure case.
A fundamental concern with respect to the mutual agreement procedure as it relates to corresponding adjustments is the failure to grant access to the mutual agreement procedure for transfer pricing cases. The undertaking to resolve by mutual agreement cases of taxation not in accordance with the Convention is an integral part of the obligations assumed by a Contracting State in entering into a tax treaty and must be performed in good faith. The failure to grant mutual agreement procedure access with respect to a treaty partner’s transfer pricing adjustments, may frustrate a primary objective of tax treaties. The work on Action 14 of the BEPS Action Plan directly addressed concerns related to the denial of access to the mutual agreement procedure with respect to a treaty partner’s transfer pricing adjustments by including, as element 1.1 of the Action 14 minimum standard, a commitment to provide access to the mutual agreement procedure in transfer pricing cases.
The Action 14 minimum standard also comprises a number of other elements intended to address more generally concerns related to the denial of access to the mutual agreement procedure. These include: a commitment to provide access to the mutual agreement procedure in cases in which there is a disagreement between the taxpayer and the tax authorities making an adjustment as to whether the conditions for the application of a treaty anti-abuse provision have been met or as to whether the application of a domestic law anti-abuse provision is in conflict with the provisions of a treaty (element 1.2); a commitment to publish rules, guidelines and procedures regarding the mutual agreement procedure (element 2.1) and to identify in that guidance the specific information and documentation that a taxpayer is required to submit with a request for mutual agreement procedure assistance (element 3.2); a commitment to clarify that audit settlements between tax authorities and taxpayers do not preclude access to the mutual agreement procedure (element 2.6); and a commitment to ensure that both competent authorities are made aware of requests for mutual agreement procedure assistance by either (i) amending Article 25(1) to permit requests to be made to the competent authority of either Contracting State or (ii) implementing a bilateral notification or consultation process for cases in which the competent authority to whom the case is presented does not consider the taxpayer’s objection to be justified (element 3.1).
Relief under paragraph 2 of Article 9 may be unavailable if the time limit provided by treaty or domestic law for making corresponding adjustments has expired. Paragraph 2 of Article 9 does not specify whether there should be a time limit after which corresponding adjustments should not be made. Some countries prefer an open-ended approach so that double taxation may be mitigated. Other countries consider the open-ended approach to be unreasonable for administrative purposes. Thus, relief may depend on whether the applicable treaty overrides domestic time limitations, establishes other time limits, or links the implementation of relief to the time limits prescribed by domestic law.
Time limits for finalising a taxpayer’s tax liability are necessary to provide certainty for taxpayers and tax administrations. In a transfer pricing case a country may under its domestic law be legally unable to make a corresponding adjustment if the time has expired for finalising the tax liability of the relevant associated enterprise. Thus, the existence of such time limits and the fact that they vary from country to country should be considered in order to minimise double taxation.
Paragraph 2 of Article 25 of the OECD Model Tax Convention addresses the time limit issue by requiring that any agreement reached by the competent authorities pursuant to the mutual agreement procedure shall be implemented notwithstanding the time limits in the domestic law of the Contracting States. Paragraph 29 of the Commentary on Article 25 recognises that the last sentence of Article 25(2) unequivocally states the obligation to implement such agreements (and notes that impediments to implementation that exist at the time a tax treaty is entered into should generally be built into the terms of the agreement itself). Time limits therefore do not impede the making of corresponding adjustments where a bilateral treaty includes this provision. Some countries, however, may be unwilling or unable to override their domestic time limits in this way and have entered explicit reservations on this point. OECD member countries therefore are encouraged as far as possible to extend domestic time limits for purposes of making corresponding adjustments when mutual agreement procedures have been invoked.
Where a bilateral treaty does not override domestic time limits for the purposes of the mutual agreement procedure, tax administrations should be ready to initiate discussions quickly upon the taxpayer’s request, well before the expiration of any time limits that would preclude the making of an adjustment. Furthermore, OECD member countries are encouraged to adopt domestic law that would allow the suspension of time limits on determining tax liability until the discussions have been concluded.
The work on Action 14 of the BEPS Action Plan directly addresses the obstacle that domestic law time limits may present to effective mutual agreement procedures. Element 3.3 of the Action 14 minimum standard includes a recommendation that countries should include the second sentence of paragraph 2 of Article 25 in their tax treaties to ensure that domestic law time limits (1) do not prevent the implementation of competent authority mutual agreements and (2) do not thereby frustrate the objective of resolving cases of taxation not in accordance with the Convention.
Where a country cannot include the second sentence of paragraph 2 of Article 25 in its tax treaties, element 3.3 of the Action 14 minimum standard states that it should be willing to accept an alternative treaty provision that limits the time during which a Contracting State may make an adjustment pursuant to Article 9(1), in order to avoid late adjustments with respect to which mutual agreement procedure relief will not be available. Such a country would satisfy this element of the minimum standard where the alternative treaty provision was drafted to reflect the time limits for adjustments provided for in that country’s domestic law. That alternative provision, as presented in the Report on BEPS Action 14, reads as follows: [In Article 9]: 3. A Contracting State shall not include in the profits of an enterprise, and tax accordingly, profits that would have accrued to the enterprise but by reason of the conditions referred to in paragraph 1 have not so accrued, after [bilaterally agreed period] from the end of the taxable year in which the profits would have accrued to the enterprise. The provisions of this paragraph shall not apply in the case of fraud, gross negligence or willful default. Element 3.3 of the Action 14 minimum standard also states that such a country accept a similar alternative provision in Article 7 with respect to adjustments to the profits that are attributable to a permanent establishment.
While it is not possible to recommend generally a time limit on initial assessments, tax administrations are encouraged to make these assessments within their own domestic time limits without extension. If the complexity of the case or lack of cooperation from the taxpayer necessitates an extension, the extension should be made for a minimum and specified time period. Further, where domestic time limits can be extended with the agreement of the taxpayer, such an extension should be made only when the taxpayer’s consent is truly voluntary. Tax examiners are encouraged to indicate to taxpayers at an early stage their intent to make an assessment based on cross-border transfer pricing, so that the taxpayer can, if it so chooses, inform the tax administration in the other interested State, which could accordingly begin to consider the relevant issues with a view to a possible mutual agreement procedure.
Another time limit that must be considered is the three-year time limit within which a taxpayer must invoke the mutual agreement procedure under Article 25 of the OECD Model Tax Convention. The three-year period begins to run from the first notification of the action resulting in taxation not in accordance with the provisions of the Convention, which can be the time when the tax administration first notifies the taxpayer of the proposed adjustment, described as the “adjustment action” or “act of taxation”, or an earlier date as discussed at paragraphs 21-24 of the Commentary on Article 25. Although some countries consider three years too short a period for invoking the procedure, other countries consider it too long and have entered reservations on this point. The Commentary on Article 25 indicates that the time limit “must be regarded as a minimum so that Contracting States are left free to agree in their bilateral conventions upon a longer period in the interests of taxpayers”. In this regard, it should be noted that element 1.1 of the Action 14 minimum standard includes a recommendation that countries include in their tax treaties paragraphs 1 through 3 of Article 25, as interpreted in the Commentary.
The three-year time limit raises the issue of determining its starting date, which is addressed at paragraphs 21-24 of the Commentary on Article 25. In particular, paragraph 21 states that the three-year time period “should be interpreted in the way most favourable to the taxpayer”. Paragraph 22 contains guidance on the determination of the date of the act of taxation. Paragraph 23 discusses self-assessment cases. Paragraph 24 clarifies that “where it is the combination of decisions or actions taken in both Contracting States resulting in taxation not in accordance with the Convention, the time limit begins to run only from the first notification of the most recent decision or action.”
In order to minimise the possibility that time limits may prevent the mutual agreement procedure from effectively ensuring relief from or avoidance of double taxation, taxpayers should be permitted to avail themselves of the procedure at the earliest possible stage, which is as soon as an adjustment appears likely. Early competent authority consultation, before any irrevocable steps are taken by either tax administration, may ensure that there are as few procedural obstacles as possible in the way of achieving a mutually acceptable conclusion to the discussions. Some competent authorities, however, may not like to be involved at such an early stage because a proposed adjustment may not result in final action or may not trigger a claim for a corresponding adjustment. Consequently, too early an invocation of the mutual agreement process may create unnecessary work.
Once discussions under the mutual agreement procedure have commenced, the proceedings may turn out to be lengthy. The complexity of transfer pricing cases may make it difficult for the competent authorities to reach a swift resolution. Distance may make it difficult for the competent authorities to meet frequently, and correspondence is often an unsatisfactory substitute for face-to-face discussions. Difficulties also arise from differences in language, procedures, and legal and accounting systems, and these may lengthen the duration of the process. The process also may be prolonged if the taxpayer delays providing all of the information the competent authorities require for a full understanding of the transfer pricing issue or issues.
Whilst the time taken to resolve a mutual agreement procedure case may vary according to its complexity, most competent authorities endeavour to reach bilateral agreement for the resolution of a mutual agreement procedure case within 24 months. Accordingly, in order to ensure the timely, effective and efficient resolution of treaty-related disputes, the minimum standard that was adopted in the context of the work on Action 14 of the BEPS Action Plan includes a commitment to seek to resolve mutual agreement procedure cases within an average timeframe of 24 months (element 1.3). Countries’ progress toward meeting that target will be periodically reviewed on the basis of the agreed reporting framework for mutual agreement procedure statistics3 that was developed to provide a tangible measure to evaluate the effects of the implementation of the Action 14 minimum standard (see elements 1.5 and 1.6). Moreover, other elements of the Action 14 minimum standard related to the authority of staff in charge of mutual agreement processes (element 2.3), performance indicators for competent authority functions (element 2.4) and adequate competent authority resources (element 2.5) are expected to contribute to the timely resolution of mutual agreement procedure cases.
More fundamentally, the adoption in tax treaties of a mandatory binding arbitration provision similar to paragraph 5 of Article 25 to resolve issues that the competent authorities have been unable to resolve within the two year period referred to in that provision should considerably reduce the risk of lengthy mutual agreement procedures. See paragraphs 4.177-4.179.
Paragraph 1 of Article 25 of the OECD Model Tax Convention gives taxpayers the right to submit a request to initiate a mutual agreement procedure. Although the taxpayer has the right to initiate the procedure, the taxpayer has no specific right to participate in the process. It has been argued that the taxpayer also should have a right to take part in the mutual agreement procedure, including the right at least to present its case to both competent authorities, and to be informed of the progress of the discussions. It should be noted in this respect that implementation of a mutual agreement in practice is subject to the taxpayer’s acceptance. Some taxpayer representatives have suggested that the taxpayer also should have a right to be present at face-to-face discussions between the competent authorities. The purpose would be to ensure that there is no misunderstanding by the competent authorities of the facts and arguments that are relevant to the taxpayer’s case.
The mutual agreement procedure envisaged in Article 25 of the OECD Model Tax Convention and adopted in many bilateral agreements is not a process of litigation. While input from the taxpayer in some cases can be helpful to the procedure, it must be recalled that the mutual agreement procedure is a government-to-government process and that any taxpayer participation in that process should be subject to the discretion and mutual agreement of the competent authorities.
Outside the context of the actual discussions between the competent authorities, it is essential for the taxpayer to give the competent authorities all the information that is relevant to the issue in a timely manner. Competent authorities have limited resources and taxpayers should make every effort to facilitate the process, particularly in complex, fact- intensive transfer pricing cases in which it may be challenging for the competent authorities to develop a complete and accurate understanding of the associated enterprises’ activities. Further, because the mutual agreement procedure is fundamentally designed as a means of providing assistance to a taxpayer, competent authorities should allow taxpayers every reasonable opportunity to present the relevant facts and arguments to them to ensure as far as possible that the matter is not subject to misunderstanding.
In practice, the competent authorities of many OECD member countries routinely give taxpayers such opportunities, keep them informed of the progress of the discussions, and often ask them during the course of the discussions whether they can accept the settlements contemplated by the competent authorities. These practices, already standard procedure in most countries, should be adopted as widely as possible. They are reflected in the OECD’s Manual for Effective Mutual Agreement Procedures.
Taxpayers’ contributions to the mutual agreement procedure process are of course facilitated where public guidance on applicable procedures is readily accessible. The work on Action 14 of the BEPS Action Plan directly recognised the importance of providing such guidance. Element 2.1 of the Action 14 minimum standard states that countries should develop and publish rules, guidelines and procedures regarding the mutual agreement procedure and take appropriate measures to make such information available to taxpayers. Such guidance should include information on how taxpayers may make requests for competent authority assistance. It should be drafted in clear and plain language and should be readily available to the public. The Report on BEPS Action 14 also notes that such information may be of particular relevance where an adjustment may potentially involve issues within the scope of a tax treaty, such as where a transfer pricing adjustment is made with respect to a controlled transaction with an associated enterprise in a treaty partner jurisdiction, and that countries should appropriately seek to ensure that mutual agreement procedure programme guidance is available to taxpayers in such cases. To promote the transparency and dissemination of such published guidance, element 2.2 of the Action 14 minimum standard includes the publication of country mutual agreement procedure profiles on a shared public platform, in order to make broadly available competent authority contact details, links to relevant domestic guidance and other useful country-specific information. These country profiles, prepared by the members of the Inclusive Framework on BEPS4 pursuant to an agreed reporting template developed for that purpose, are published on the OECD website.
The work on Action 14 also addresses a number of other aspects related to the content of mutual agreement procedure programme guidance: Element 3.2 of the Action 14 minimum standard states that countries should identify in their mutual agreement procedure programme guidance the specific information and documentation that a taxpayer is required to submit with a request for competent authority assistance. Pursuant to element 3.2, countries should not deny access to the mutual agreement procedure based on the argument that a taxpayer has provided insufficient information where the taxpayer has provided the required information and documentation consistent with such guidance. Element 2.6 of the Action 14 minimum standard states that countries should clarify in their mutual agreement procedure programme guidance that audit settlements between tax authorities and taxpayers do not preclude access to the mutual agreement procedure. Certain of the non-binding Action 14 best practices additionally recommend that countries’ mutual agreement procedure programme guidance should include: an explanation of the relationship between the mutual agreement procedure and domestic law administrative and judicial remedies (best practice 8); guidance on the consideration of interest and penalties in the mutual agreement procedure (best practice 10); and guidance on multilateral mutual agreement procedures and advance pricing arrangements (best practice 11). Best practice 9 recommends that this guidance provide that taxpayers will be allowed access to the mutual agreement procedure so that the competent authorities can resolve through consultation the double taxation that can arise in the case of bona fide taxpayer initiated foreign adjustments.
There is no need for the competent authorities to agree to rules or guidelines governing the procedure, since the rules or guidelines would be limited in effect to the competent authority’s relationship with taxpayers seeking its assistance. However, competent authorities should routinely communicate such unilateral rules or guidelines to the competent authorities of their treaty partners and ensure that their country mutual agreement procedure profiles (see paragraph 4.62 above) are kept up-to-date.
The process of obtaining relief from double taxation through a corresponding adjustment can be complicated by issues relating to the collection of tax deficiencies and the assessment of interest on those deficiencies or overpayment. A first problem is that the assessed deficiency may be collected before the corresponding adjustment proceeding is completed, because of a lack of domestic procedures allowing the collection to be suspended. This may cause the MNE group to pay the same tax twice until the issues can be resolved. This problem arises not only in the context of the mutual agreement procedure but also for internal appeals.. The work on Action 14 of the BEPS Action Plan recognised that the collection of tax by both Contracting States pending the resolution of a case through the mutual agreement procedure may have a significant impact on a taxpayer’s business (for example, as a result of cash flow problems). Such collection of tax may also make it more difficult for a competent authority to engage in good faith mutual agreement procedure discussions when it considers that it may likely have to refund taxes already collected. The Report on BEPS Action 14 accordingly includes as best practice 6 a recommendation that countries should take appropriate measures to provide for a suspension of collection procedures during the period in which a mutual agreement procedure case is pending; such a suspension of collections should be available, at a minimum, under the same conditions as apply to a person pursuing a domestic administrative or judicial remedy. In this regard, it should be noted that the country mutual agreement procedure profiles prepared pursuant to element 2.2 of the Action 14 minimum standard (see paragraph 4.62) include information on the availability of procedures for the suspension of collections in specific countries.
Whether or not collection of the deficiency is suspended or partially suspended, other complications may arise. Because of the lengthy time period for processing many transfer pricing cases, the interest due on a deficiency or, if a corresponding adjustment is allowed, on the overpayment of tax in the other country can equal or exceed the amount of the tax itself. Countries should take into account in their mutual agreement procedures that inconsistent interest rules across the two jurisdictions may result in additional cost for the MNE group, or in other cases provide a benefit to the MNE group (e.g. where the interest paid in the country making the corresponding adjustment exceeds the interest imposed in the country making the primary adjustment) that would not have been available if the controlled transactions had been undertaken on an arm’s length basis originally. As noted above, the Report on BEPS Action 14 includes as best practice 10 a recommendation that countries’ published mutual agreement procedure guidance should provide guidance on the consideration of interest in the mutual agreement procedure. In addition, the country mutual agreement procedure profiles prepared pursuant to element 2.2 of the Action 14 minimum standard include information on how interest and penalties are dealt with by specific countries in the context of the mutual agreement procedure.
The amount of interest (as distinct from the rate at which it is applied) may also have more to do with the year to which the jurisdiction making the corresponding adjustment attributes the corresponding adjustment. The jurisdiction making the corresponding adjustment may decide to make the adjustment for the year in which the primary adjustment is determined, in which case relatively little interest is likely to be payable (regardless of the rate of interest), whereas the jurisdiction making the primary adjustment may seek to impose interest on the understated and uncollected tax liability from the year in which the controlled transactions took place (notwithstanding that a relatively low rate of interest may be imposed). The issue of the year to which a corresponding adjustment is attributed is raised in paragraph 4.36. It may be appropriate in certain cases for both competent authorities to agree not to assess or pay interest in connection with the adjustment at issue, but this may not be possible in the absence of a specific provision addressing this issue in the relevant bilateral treaty. This approach would also reduce administrative complexities. However, as the interest on the deficiency and the interest on the overpayment are attributable to different taxpayers in different jurisdictions, there would be no assurance under such an approach that a proper economic result would be achieved.
Corresponding adjustments are not the only adjustments that may be triggered by a primary transfer pricing adjustment. Primary transfer pricing adjustments and their corresponding adjustments change the allocation of taxable profits of an MNE group for tax purposes but they do not alter the fact that the excess profits represented by the adjustment are not consistent with the result that would have arisen if the controlled transactions had been undertaken on an arm’s length basis. To make the actual allocation of profits consistent with the primary transfer pricing adjustment, some countries having proposed a transfer pricing adjustment will assert under their domestic legislation a constructive transaction (a secondary transaction), whereby the excess profits resulting from a primary adjustment are treated as having been transferred in some other form and taxed accordingly. Ordinarily, the secondary transactions will take the form of constructive dividends, constructive equity contributions, or constructive loans. For example, a country making a primary adjustment to the income of a subsidiary of a foreign parent may treat the excess profits in the hands of the foreign parent as having been transferred as a dividend, in which case withholding tax may apply. It may be that the subsidiary paid an excessive transfer price to the foreign parent as a means of avoiding that withholding tax. Thus, secondary adjustments attempt to account for the difference between the re-determined taxable profits and the originally booked profits. The subjecting to tax of a secondary transaction gives rise to a secondary transfer pricing adjustment (a secondary adjustment). Thus, secondary adjustments may serve to prevent tax avoidance. The exact form that a secondary transaction takes and of the consequent secondary adjustment will depend on the facts of the case and on the tax laws of the country that asserts the secondary adjustment.
Another example of a tax administration seeking to assert a secondary transaction may be where the tax administration making a primary adjustment treats the excess profits as being a constructive loan from one associated enterprise to the other associated enterprise. In this case, an obligation to repay the loan would be deemed to arise. The tax administration making the primary adjustment may then seek to apply the arm’s length principle to this secondary transaction to impute an arm’s length rate of interest. The interest rate to be applied, the timing to be attached to the making of interest payments, if any, and whether interest is to be capitalised would generally need to be addressed. The constructive loan approach may have an effect not only for the year to which a primary adjustment relates but to subsequent years until such time as the constructive loan is considered by the tax administration asserting the secondary adjustment to have been repaid.
A secondary adjustment may result in double taxation unless a corresponding credit or some other form of relief is provided by the other country for the additional tax liability that may result from a secondary adjustment. Where a secondary adjustment takes the form of a constructive dividend any withholding tax which is then imposed may not be relievable because there may not be a deemed receipt under the domestic legislation of the other country.
The Commentary on paragraph 2 of Article 9 of the OECD Model Tax Convention notes that the Article does not deal with secondary adjustments, and thus it neither forbids nor requires tax administrations to make secondary adjustments. In a broad sense, the purpose of double tax agreements can be stated as being for the avoidance of double taxation and the prevention of fiscal evasion with respect to taxes on income and capital. Many countries do not make secondary adjustments either as a matter of practice or because their respective domestic provisions do not permit them to do so. Some countries might refuse to grant relief in respect of other countries’ secondary adjustments and indeed they are not required to do so under Article 9.
Secondary adjustments are rejected by some countries because of the practical difficulties they present. For example, if a primary adjustment is made between brother-sister companies, the secondary adjustment may involve a hypothetical dividend from one of those companies up a chain to a common parent, followed by constructive equity contributions down another chain of ownership to reach the other company involved in the transaction. Many hypothetical transactions might be created, raising questions whether tax consequences should be triggered in other jurisdictions besides those involved in the transaction for which the primary adjustment was made. This might be avoided if the secondary transaction were a loan, but constructive loans are not used by most countries for this purpose and they carry their own complications because of issues relating to imputed interest. It would be inappropriate for minority shareholders that are not parties to the controlled transactions and that have accordingly not received excess cash to be considered recipients of a constructive dividend, even though a non-pro- rata dividend might be considered inconsistent with the requirements of applicable corporate law. In addition, as a result of the interaction with the foreign tax credit system, a secondary adjustment may excessively reduce the overall tax burden of the MNE group.
In light of the foregoing difficulties, tax administrations, when secondary adjustments are considered necessary, are encouraged to structure such adjustments in a way that the possibility of double taxation as a consequence thereof would be minimised, except where the taxpayer’s behaviour suggests an intent to disguise a dividend for purposes of avoiding withholding tax. In addition, countries in the process of formulating or reviewing policy on this matter are recommended to take into consideration the above-mentioned difficulties.
Some countries that have adopted secondary adjustments also give the taxpayer receiving the primary adjustment another option that allows the taxpayer to avoid the secondary adjustment by having the taxpayer arrange for the MNE group of which it is a member to repatriate the excess profits to enable the taxpayer to conform its accounts to the primary adjustment. The repatriation could be effected either by setting up an account receivable or by reclassifying other transfers, such as dividend payments where the adjustment is between parent and subsidiary, as a payment of additional transfer price (where the original price was too low) or as a refund of transfer price (where the original price was too high).
Where a repatriation involves reclassifying a dividend payment, the amount of the dividend (up to the amount of the primary adjustment) would be excluded from the recipient’s gross income (because it would already have been accounted for through the primary adjustment). The consequences would be that the recipient would lose any indirect tax credit (or benefit of a dividend exemption in an exemption system) and a credit for withholding tax that had been allowed on the dividend.
When the repatriation involves establishing an account receivable, the adjustments to actual cash flow will be made over time, although domestic law may limit the time within which the account can be satisfied. This approach is identical to using a constructive loan as a secondary transaction to account for excess profits in the hands of one of the parties to the controlled transaction. The accrual of interest on the account could have its own tax consequences, however, and this may complicate the process, depending upon when interest begins to accrue under domestic law (as discussed in paragraph 4.69). Some countries may be willing to waive the interest charge on these accounts as part of a competent authority agreement.
Where a repatriation is sought, a question arises about how such payments or arrangements should be recorded in the accounts of the taxpayer repatriating the payment to its associated enterprise so that both it and the tax administration of that country are aware that a repatriation has occurred or has been set up. The actual recording of the repatriation in the accounts of the enterprise from whom the repatriation is sought will ultimately depend on the form the repatriation takes. For example, where a dividend receipt is to be regarded by the tax administration making the primary adjustment and the taxpayer receiving the dividend as the repatriation, then this type of arrangement may not need to be specially recorded in the accounts of the associated enterprise paying the dividend, as such an arrangement may not affect the amount or characterisation of the dividend in its hands. On the other hand, where an account payable is set up, both the taxpayer recording the account payable and the tax administration of that country will need to be aware that the account payable relates to a repatriation so that any repayments from the account or of interest on the outstanding balance in the account are clearly able to be identified and treated according to the domestic laws of that country. In addition, issues may be presented in relation to currency exchange gains and losses.
As most OECD member countries at this time have not had much experience with the use of repatriation, it is recommended that agreements between taxpayers and tax administrations for a repatriation to take place be discussed in the mutual agreement proceeding where it has been initiated for the related primary adjustment.
A simultaneous tax examination is a form of mutual assistance, used in a wide range of international issues, that allows two or more countries to cooperate in tax investigations. Simultaneous tax examinations can be particularly useful where information based in a third country is a key to a tax investigation, since they generally lead to more timely and more effective exchanges of information. Historically, simultaneous tax examinations of transfer pricing issues have focused on cases where the true nature of transactions was obscured by the interposition of tax havens. However, in complex transfer pricing cases, it is suggested that simultaneous examinations could serve a broader role since they may improve the adequacy of data available to the participating tax administrations for transfer pricing analyses. It has also been suggested that simultaneous examinations could help reduce the possibilities for economic double taxation, reduce the compliance cost to taxpayers, and speed up the resolution of issues. In a simultaneous examination, if a reassessment is made, both countries involved should endeavour to reach a result that avoids double taxation for the MNE group.
Simultaneous tax examinations are defined in Part A of the OECD Model Agreement for the Undertaking of Simultaneous Tax Examinations (OECD Model Agreement). According to this agreement, a simultaneous tax examination means an “arrangement between two or more parties to examine simultaneously and independently, each on its own territory, the tax affairs of (a) taxpayer(s) in which they have a common or related interest with a view to exchanging any relevant information which they so obtain”. This form of mutual assistance is not meant to be a substitute for the mutual agreement procedure. Any exchange of information as a result of the simultaneous tax examination continues to be exchanged via the competent authorities, with all the safeguards that are built into such exchanges. Practical information on simultaneous examinations can be found in the relevant module of the Manual on Information Exchange that was adopted by the Committee on Fiscal Affairs on 23 January 2006 (see http://www.oecd.org/ctp/eoi/manual).
While provisions that follow Article 26 of the OECD Model Tax Convention may provide the legal basis for conducting simultaneous examinations, competent authorities frequently conclude working arrangements that lay down the objectives of their simultaneous tax examination programs and practical procedures connected with the simultaneous tax examination and exchange of information. Once such an agreement has been reached on the general lines to be followed and specific cases have been selected, tax examiners and inspectors of each state will separately carry out their examination within their own jurisdiction and pursuant to their domestic law and administrative practice.
Simultaneous tax examinations are within the scope of the exchange of information provision based on Article 26 of the OECD Model Tax Convention. Article 26 provides for cooperation between the competent authorities of the Contracting States in the form of exchanges of information necessary for carrying out the provisions of the Convention or of their domestic laws concerning taxes covered by the Convention. Article 26 and the Commentary do not restrict the possibilities of assistance to the three methods of exchanging information mentioned in the Commentary (exchange on request, spontaneous exchanges, and automatic exchanges).
Simultaneous tax examinations may be authorised outside the context of double tax treaties. For example, Article 12 of the Nordic Convention on Mutual Assistance in Tax Matters governs exchange of information and assistance in tax collection between the Nordic countries and provides for the possibility of simultaneous tax examinations. This convention gives common guidelines for the selection of cases and for carrying out such examinations. Article 8 of the joint Council of Europe and OECD Convention on Mutual Administrative Assistance in Tax Matters also provides expressly for the possibility of simultaneous tax examinations.
In all cases the information obtained by the tax administration of a state has to be treated as confidential under its domestic legislation and may be used only for certain tax purposes and disclosed only to certain persons and authorities involved in specifically defined tax matters covered by the tax treaty or mutual assistance agreement. The taxpayers affected are normally notified of the fact that they have been selected for a simultaneous examination and in some countries they may have the right to be informed when the tax administrations are considering a simultaneous tax examination or when information will be transmitted in conformity with Article 26. In such cases, the competent authority should inform its counterpart in the foreign state that such disclosure will occur.
In selecting transfer pricing cases for simultaneous examinations, there may be major obstacles caused by the differences in time limits for conducting examinations or making assessments in different countries and the different tax periods open for examination. However, these problems may be mitigated by an early exchange of examination schedules between the relevant competent authorities to find out in which cases the tax examination periods coincide and to synchronise future examination periods. While at first glance an early exchange of examination schedules would seem beneficial, some countries have found that the chances of a treaty partner accepting a proposal are considerably better when one is able to present issues more comprehensively to justify a simultaneous examination.
Once a case is selected for a simultaneous examination it is customary for tax inspectors or examiners to meet, to plan, to coordinate and to follow closely the progress of the simultaneous tax examination. Especially in complex cases, meetings of the tax inspectors or examiners concerned may also be held with taxpayer participation to clarify factual issues. In those countries where the taxpayer has the right to be consulted before information is transferred to another tax administration, this procedure should also be followed in the context of a simultaneous examination. In this situation, that tax administration should inform in advance its treaty partners that it is subject to this requirement before the simultaneous examination is begun.
Simultaneous tax examinations may be a useful instrument to determine the correct tax liability of associated enterprises in cases where, for example, costs are shared or charged and profits are allocated between taxpayers in different taxing jurisdictions or more generally where transfer pricing issues are involved. Simultaneous tax examinations may facilitate an exchange of information on multinational business practices, complex transactions, cost contribution arrangements, and profit allocation methods in special fields such as global trading and innovative financial transactions. As a result, tax administrations may acquire a better understanding of and insight into the overall activities of an MNE and obtain extended possibilities of comparison and checking international transactions. Simultaneous tax examinations may also support the industry-wide exchange of information, which is aimed at developing knowledge of taxpayer behaviour, practices and trends within an industry, and other information that might be suitable beyond the specific cases examined.
One objective of simultaneous tax examinations is to promote compliance with transfer pricing regulations. Obtaining the necessary information and determining the facts and circumstances about such matters as the transfer pricing conditions of controlled transactions between associated enterprises in two or more tax jurisdictions may be difficult for a tax administration, especially in cases where the taxpayer in its jurisdiction does not cooperate or fails to provide the necessary information in due time. The simultaneous tax examination process can help tax administrations to establish these facts faster and more effectively and economically.
The process also might allow for the identification of potential transfer pricing disputes at an early stage, thereby minimising litigation with taxpayers. This could happen when, based upon the information obtained in the course of a simultaneous tax examination, the participating tax examiners or inspectors have the opportunity to discuss any differences in opinion with regard to the transfer pricing conditions which exist between the associated enterprises and are able to reconcile these contentions. When such a process is undertaken, the tax examiners or inspectors concerned should, as far as possible, arrive at concurring statements as to the determination and evaluation of the facts and circumstances of the controlled transactions between the associated enterprises, stating any disagreements about the evaluation of facts, and any differences with respect to the legal treatment of the transfer pricing conditions which exist between the associated enterprises. Such statements could then serve as a basis for subsequent mutual agreement procedures and perhaps obviate the problems caused by one country examining the affairs of a taxpayer long after the treaty partner country has finally settled the tax liability of the relevant associated enterprise. For example, such an approach could minimise mutual agreement procedure difficulties due to the lack of relevant information.
In some cases the simultaneous tax examination procedure may allow the participating tax administrations to reach an agreement on the transfer pricing conditions of a controlled transaction between the associated enterprises. Where an agreement is reached, corresponding adjustments may be made at an early stage, thus avoiding time-limit impediments and economic double taxation to the extent possible. In addition, if the agreement about the associated enterprises’ transfer pricing is reached with the taxpayers’ consent, time-consuming and expensive litigation may be avoided.
Even if no agreement between the tax administrations can be reached in the course of a simultaneous tax examination with respect to the associated enterprises’ transfer pricing, the OECD Model Agreement envisions that either associated enterprise may be able to present a request for the opening of a mutual agreement procedure to avoid economic double taxation at an earlier stage than it would have if there were no simultaneous tax examination. If this is the case, then simultaneous tax examinations may significantly reduce the time span between a tax administration’s adjustments made to a taxpayer’s tax liability and the implementation of a mutual agreement procedure. Moreover, the OECD Model Agreement envisions that simultaneous tax examinations may facilitate mutual agreement procedures, because tax administrations will be able to build up more complete factual evidence for those tax adjustments for which a mutual agreement procedure may be requested by a taxpayer. Based upon the determination and evaluation of facts and the proposed tax treatment of the transfer pricing issues concerned as set forth in the tax administrations’ statements described above, the practical operation of the mutual agreement procedure may be improved significantly, allowing the competent authorities to reach an agreement more easily.
The associated enterprises may also benefit from simultaneous tax examinations from the savings of time and resources due to the coordination of inquiries from the tax administrations involved and the avoidance of duplication. In addition, the simultaneous involvement of two or more tax administrations in the examination of transfer pricing between associated enterprises may provide the opportunity for an MNE to take a more active role in resolving its transfer pricing issues. By presenting the relevant facts and arguments to each of the participating tax administrations during the simultaneous tax examination the associated enterprises may help avoid misunderstandings and facilitate the tax administrations’ concurring determination and evaluation of their transfer pricing conditions. Thus, the associated enterprises may obtain certainty with regard to their transfer pricing at an early stage. See paragraph 4.79.
As a result of the increased use of simultaneous tax examinations among OECD member countries, the Committee on Fiscal Affairs decided it would be useful to draft the OECD Model Agreement for those countries that are able and wish to engage in this type of cooperation. On 23 July 1992, the Council of the OECD made a recommendation to member countries to use this Model Agreement, which provides guidelines on the legal and practical aspects of this form of cooperation.
With the increasing internationalisation of trade and business and the complexity of transactions of MNEs, transfer pricing issues have become more and more important. Simultaneous tax examinations can alleviate the difficulties experienced by both taxpayers and tax administrations connected with the transfer pricing of MNEs. A greater use of simultaneous tax examinations is therefore recommended in the examination of transfer pricing cases and to facilitate exchange of information and the operation of mutual agreement procedures. In a simultaneous examination, if a reassessment is made, both countries involved should endeavour to reach a result that avoids double taxation for the MNE group.
Applying the arm’s length principle can be a resource-intensive process. It may impose a heavy administrative burden on taxpayers and tax administrations that can be exacerbated by both complex rules and resulting compliance demands. These facts have led OECD member countries to consider whether and when safe harbour rules would be appropriate in the transfer pricing area.
When these Guidelines were adopted in 1995, the view expressed regarding safe harbour rules was generally negative. It was suggested that while safe harbours could simplify transfer pricing compliance and administration, safe harbour rules may raise fundamental problems that could potentially have perverse effects on the pricing decisions of enterprises engaged in controlled transactions. It was suggested that unilateral safe harbours may have a negative impact on the tax revenues of the country implementing the safe harbour, as well as on the tax revenues of countries whose associated enterprises engage in controlled transactions with taxpayers electing a safe harbour. It was further suggested that safe harbours may not be compatible with the arm’s length principle. Therefore, it was concluded that transfer pricing safe harbours are not generally advisable, and consequently the use of safe harbours was not recommended.
Despite these generally negative conclusions, a number of countries have adopted safe harbour rules. Those rules have generally been applied to smaller taxpayers and/or less complex transactions. They are generally evaluated favourably by both tax administrations and taxpayers, who indicate that the benefits of safe harbours outweigh the related concerns when such rules are carefully targeted and prescribed and when efforts are made to avoid the problems that could arise from poorly considered safe harbour regimes.
The appropriateness of safe harbours can be expected to be most apparent when they are directed at taxpayers and/or transactions which involve low transfer pricing risks and when they are adopted on a bilateral or multilateral basis. It should be recognised that a safe harbour provision does not bind or limit in any way any tax administration other than the tax administration that has expressly adopted the safe harbour.
Although safe harbours primarily benefit taxpayers, by providing for a more optimal use of resources, they can benefit tax administrations as well. Tax administrations can shift audit and examination resources from smaller taxpayers and less complex transactions (which may typically be resolved in practice on a consistent basis as to both transfer pricing methodology and actual results) to more complex, higher-risk cases. At the same time, taxpayers can price eligible transactions and file their tax returns with more certainty and with lower compliance burdens. However, the design of safe harbours requires careful attention to concerns about the degree of approximation to arm’s length prices that would be permitted in determining transfer prices under safe harbour rules for eligible taxpayers, the potential for creating inappropriate tax planning opportunities including double non-taxation of income, equitable treatment of similarly situated taxpayers, and the potential for double taxation resulting from the possible incompatibility of the safe harbours with the arm’s length principle or with the practices of other countries.
The following discussion considers the benefits of, and concerns regarding, safe harbour provisions and provides guidance regarding the circumstances in which safe harbours may be applied in a transfer pricing system based on the arm’s length principle.
Some of the difficulties that arise in applying the arm’s length principle may be avoided by providing circumstances in which eligible taxpayers may elect to follow a simple set of prescribed transfer pricing rules in connection with clearly and carefully defined transactions, or may be exempted from the application of the general transfer pricing rules. In the former case, prices established under such rules would be automatically accepted by the tax administrations that have expressly adopted such rules. These elective provisions are often referred to as “safe harbours”.
A safe harbour in a transfer pricing regime is a provision that applies to a defined category of taxpayers or transactions and that relieves eligible taxpayers from certain obligations otherwise imposed by a country’s general transfer pricing rules. A safe harbour substitutes simpler obligations for those under the general transfer pricing regime. Such a provision could, for example, allow taxpayers to establish transfer prices in a specific way, e.g. by applying a simplified transfer pricing approach provided by the tax administration. Alternatively, a safe harbour could exempt a defined category of taxpayers or transactions from the application of all or part of the general transfer pricing rules. Often, eligible taxpayers complying with the safe harbour provision will be relieved from burdensome compliance obligations, including some or all associated transfer pricing documentation requirements.
For purposes of the discussion in this Section, safe harbours do not include administrative simplification measures which do not directly involve determination of arm’s length prices, e.g. simplified, or exemption from, documentation requirements (in the absence of a pricing determination), and procedures whereby a tax administration and a taxpayer agree on transfer pricing in advance of the controlled transactions (advance pricing arrangements), which are discussed in Section F of this chapter. The discussion in this section also does not extend to tax provisions designed to prevent “excessive” debt in a foreign subsidiary (“thin capitalisation” rules).
Although they would not fully meet the foregoing description of a safe harbour, it may be the case that some countries adopt other administrative simplification measures that use presumptions to realise some of the benefits discussed in this Section. For example, a rebuttable presumption might be established under which a mandatory pricing target would be established by a tax authority, subject to a taxpayer’s right to demonstrate that its transfer price is consistent with the arm’s length principle. Under such a system, it would be essential that the taxpayer does not bear a higher burden to demonstrate its price is consistent with the arm’s length principle than it would if no such system were in place. In any such system, it would be essential to permit resolution of cases of double taxation arising from application of the mandatory presumption through the mutual agreement process.
The basic benefits of safe harbours are as follows: Simplifying compliance and reducing compliance costs for eligible taxpayers in determining and documenting appropriate conditions for qualifying controlled transactions; Providing certainty to eligible taxpayers that the price charged or paid on qualifying controlled transactions will be accepted by the tax administrations that have adopted the safe harbour with a limited audit or without an audit beyond ensuring the taxpayer has met the eligibility conditions of, and complied with, the safe harbour provisions; Permitting tax administrations to redirect their administrative resources from the examination of lower risk transactions to examinations of more complex or higher risk transactions and taxpayers.
Application of the arm’s length principle may require collection and analysis of data that may be difficult or costly to obtain and/or evaluate. In certain cases, such compliance burdens may be disproportionate to the size of the taxpayer, its functions performed, and the transfer pricing risks inherent in its controlled transactions.
Properly designed safe harbours may significantly ease compliance burdens by eliminating data collection and associated documentation requirements in exchange for the taxpayer pricing qualifying transactions within the parameters set by the safe harbour. Especially in areas where transfer pricing risks are small, and the burden of compliance and documentation is disproportionate to the transfer pricing exposure, such a trade-off may be mutually advantageous to taxpayers and tax administrations. Under a safe harbour, taxpayers would be able to establish transfer prices which will not be challenged by tax administrations providing the safe harbour without being obligated to search for comparable transactions or expend resources to demonstrate transfer pricing compliance to such tax administrations.
Another advantage provided by a safe harbour is the certainty that the taxpayer’s transfer prices will be accepted by the tax administration providing the safe harbour, provided they have met the eligibility conditions of, and complied with, the safe harbour provisions. The tax administration would accept, with limited or no scrutiny, transfer prices within the safe harbour parameters. Taxpayers could be provided with relevant parameters which would provide a transfer price deemed appropriate by the tax administration for the qualifying transaction.
A safe harbour would result in a degree of administrative simplicity for the tax administration. Although the eligibility of particular taxpayers or transactions for the safe harbour would need to be carefully evaluated, depending on the specific safe harbour provision, such evaluations would not necessarily have to be performed by auditors with transfer pricing expertise. Once eligibility for the safe harbour has been established, qualifying taxpayers would require minimal examination with respect to the transfer prices of controlled transactions qualifying for the safe harbour. This would enable tax administrations to secure tax revenues in low risk situations with a limited commitment of administrative resources and to concentrate their efforts on the examination of more complex or higher risk transactions and taxpayers. A safe harbour may also increase the level of compliance among small taxpayers that may otherwise believe their transfer pricing practices will escape scrutiny.
The availability of safe harbours for a given category of taxpayers or transactions may have adverse consequences. These concerns stem from the fact that: The implementation of a safe harbour in a given country may lead to taxable income being reported that is not in accordance with the arm’s length principle; Safe harbours may increase the risk of double taxation or double non-taxation when adopted unilaterally; Safe harbours potentially open avenues for inappropriate tax planning, and Safe harbours may raise issues of equity and uniformity.
Where a safe harbour provides a simplified transfer pricing approach, it may not correspond in all cases to the most appropriate method applicable to the facts and circumstances of the taxpayer under the general transfer pricing provisions. For example, a safe harbour might require the use of a particular method when the taxpayer could otherwise have determined that another method was the most appropriate method under the facts and circumstances. Such an occurrence could be considered as inconsistent with the arm’s length principle, which requires the use of the most appropriate method.
Safe harbours involve a trade-off between strict compliance with the arm’s length principle and administrability. They are not tailored to fit exactly the varying facts and circumstances of individual taxpayers and transactions. The degree of approximation of prices determined under the safe harbour with prices determined in accordance with the arm’s length principle could be improved by collecting, collating, and frequently updating a pool of information regarding prices and pricing developments in respect of the relevant types of transactions between uncontrolled parties of the relevant nature. However, such efforts to set safe harbour parameters accurately enough to satisfy the arm’s length principle could erode the administrative simplicity of the safe harbour.
Any potential disadvantages to taxpayers from safe harbours diverging from arm’s length pricing are avoided when taxpayers have the option to either elect the safe harbour or price transactions in accordance with the arm’s length principle. With such an approach, taxpayers that believe the safe harbour would require them to report an amount of income exceeding the arm’s length amount could apply the general transfer pricing rules. While such an approach can limit the divergence from arm’s length pricing under a safe harbour regime, it would also limit the administrative benefits of the safe harbour to the tax administration. Moreover, tax administrations would need to consider the potential loss of tax revenue from such an approach where taxpayers will pay tax only on the lesser of the safe harbour amount or the arm’s length amount. Countries may also be concerned over the ability of taxpayers to opt in and out of a safe harbour, depending on whether the use of the safe harbour is favourable to the taxpayer in a particular year. Countries may be able to gain greater comfort regarding this risk by controlling the conditions under which a taxpayer can be eligible for the safe harbour, for example by requiring taxpayers to notify the tax authority in advance of using the safe harbour or to commit to its use for a certain number of years.
One major concern raised by a safe harbour is that it may increase the risk of double taxation. If a tax administration sets safe harbour parameters at levels either above or below arm’s length prices in order to increase reported profits in its country, it may induce taxpayers to modify the prices that they would otherwise have charged or paid to controlled parties, in order to avoid transfer pricing scrutiny in the safe harbour country. The concern of possible overstatement of taxable income in the country providing the safe harbour is greater where that country imposes significant penalties for understatement of tax or failure to meet documentation requirements, with the result that there may be added incentives to ensure that the transfer pricing is accepted in that country without further review.
If the safe harbour causes taxpayers to report income above arm’s length levels, it would work to the benefit of the tax administration providing the safe harbour, as more taxable income would be reported by such domestic taxpayers. On the other hand, the safe harbour may lead to less taxable income being reported in the tax jurisdiction of the foreign associated enterprise that is the other party to the transaction. The other tax administrations may then challenge prices derived from the application of a safe harbour, with the result that the taxpayer would face the prospect of double taxation. Accordingly, any administrative benefits gained by the tax administration of the safe harbour country would potentially be obtained at the expense of other countries which, in order to protect their own tax base, would have to determine systematically whether the prices or results permitted under the safe harbour are consistent with what would be obtained by the application of their own transfer pricing rules. The administrative burden saved by the country offering the safe harbour would therefore be shifted to the foreign jurisdictions.
In cases involving smaller taxpayers or less complex transactions, the benefits of safe harbours may outweigh the problems raised by such provisions. Provided the safe harbour is elective, taxpayers may consider that a moderate level of double taxation, if any arises because of the safe harbour, is an acceptable price to be paid in order to obtain relief from the necessity of complying with complex transfer pricing rules. One may argue that the taxpayer is capable of making its own decision in electing the safe harbour as to whether the possibility of double taxation is acceptable or not.