The impact of coronavirus (“COVID-19”) has been profound. The rapid spread of the virus has strained local medical infrastructures, led to restrictions on travel and social contact, and created unprecedented disruptions to the global economy.

During the pandemic period, many enterprises have faced or continue to face significant cash flow constraints, requiring them to develop and implement strategies to conserve and generate cash. Enterprises have seen wide swings in profitability, both upward and downward. Enterprises across a variety of industries have faced disruption to their supply chains, including the curtailment of their operations and corresponding reductions in output, and have been forced to change how their business is conducted (e.g. working from home). In many jurisdictions, factories, mines, shops and restaurants have been forced to close, at least temporarily. In some industries, demand has completely collapsed, while in others it has merely shifted channels or even increased (e.g. the market for online videoconferencing services). In the presence of significant financial hardship, some enterprises have reviewed their contractual arrangements with third parties to ascertain whether they remain bound by them or have attempted to renegotiate key terms, including requesting discounts or deferred payment. Given the significance and speed of the economic impact of the virus, governments have adopted comprehensive policy responses to support the economy and protect people’s jobs and incomes.

The arm’s length principle has been found to work effectively in the vast majority of cases,1 and this principle-based approach to assessing intercompany prices is equally robust for evaluating controlled transactions in the face of the COVID-19 pandemic. The OECD TPG are intended to help tax administrations and MNEs find mutually satisfactory solutions to transfer pricing cases2 and should be relied upon when performing a transfer pricing analysis under the possibly unique circumstances introduced by the pandemic. 1Paragraph 1.9 of Chapter I of the OECD TPG. 2 Paragraph 15 of the Preface of the OECD TPG.

However, the unique and almost unprecedented economic conditions arising from and government responses to COVID-19 have led to practical challenges for the application of the arm’s length principle. For example, the pandemic may raise novel issues or exacerbate in complexity or magnitude the occurrence of certain transfer pricing issues (e.g. effect of government assistance or the availability of reliable comparable data). For taxpayers applying transfer pricing rules for the financial years impacted by the COVID-19 pandemic and for tax administrations that will be evaluating this application, there is a need to address these practical questions. Based on the responses to the questionnaires submitted to members of the Inclusive Framework and businesses, and conscious of the need to provide practical and timely guidance, this note addresses four priority issues: (i) comparability analysis; (ii) allocation of losses and the allocation of COVID-19 specific costs; (iii) government assistance programmes; and (iv) Advance Pricing Arrangements (“APAs”). For ease of presentation, these issues have been presented as discrete topics, but it is important to emphasise that in performing a transfer pricing analysis, these topics may be interrelated and therefore should be considered together and within the analytical framework of the OECD TPG. For example, in order to determine whether an entity should be allocated losses during the pandemic under arm’s length conditions, the guidance in Chapter II of this document is relevant, but the guidance in Chapter I (as it relates to the results of the comparability analysis) and the guidance in Chapter III (if it receives government assistance) is also relevant.

It is important not to lose sight of the objective to find a reasonable estimate of an arm’s length outcome, which requires an exercise of judgment on the part of taxpayers and tax administrations.3 Accordingly, this guidance focuses on how the arm’s length principle and the OECD TPG apply to issues that may arise or be exacerbated in the context of the COVID-19 pandemic. Thus, it should be regarded as an application of existing guidance under the OECD TPG to fact patterns that may arise commonly in connection with the pandemic, and it should not be regarded as an expansion or revision of the OECD TPG, either with respect to such pandemic-related fact patterns or more generally. 3 Paragraph 1.13 of Chapter I of the OECD TPG.

This guidance acknowledges that the economic impact of the COVID-19 pandemic varies widely across economies, industries and businesses, which is a key factor when considering and interpreting its content. Therefore, in any transfer pricing analysis of the implications of the COVID-19 pandemic, businesses should seek to contemporaneously document how, and to what extent, they have been impacted by the pandemic.

The significance of risk is particularly relevant in the current economic climate for the four issues discussed in this note. The COVID-19 pandemic, which constitutes a hazard risk, has led to unusual outcomes of other risks for some taxpayers, including: (i) marketplace risk, as demand for some products and services has collapsed; (ii) operational risk, as the pandemic has disrupted supply chains and inhibited production; and (iii) financial risks, as borrowing costs for some industries have spiked and customers have delayed or defaulted on payments.4 4 Paragraph 1.72 of Chapter I of the OECD TPG.

Against this background, taxpayers and tax administrations should carefully follow the guidance on the accurate delineation of controlled transactions in Chapter I of the OECD TPG to identify with specificity the economically significant risks and to determine the specific economically significant risks that each party to a controlled transaction assumes.5 Therefore, the interplay between the COVID-19 hazard risk and other economically significant risks should be evaluated when considering risk assumption in a particular controlled transaction. In undertaking this analysis, it may be determined that a party to a controlled transaction cannot influence the hazard risk associated with a pandemic, but nevertheless assumes other risks that have materialised as a result of COVID-19. Care must also be taken to determine how the associated enterprises and the group as a whole respond to the manifestation of hazard risks and its subsequent effects on the other economically significant risks identified in the controlled transaction. (See paragraphs 1.34 and 1.35 of Chapter I of the OECD TPG). In particular, the widespread effects of the COVID-19 pandemic in an industry or within an MNE group do not suffice to claim that a member of an MNE group has to bear the consequences of risks materialising as a result of the COVID-19 pandemic without an analysis of how the outcome of the economically significant risks controlled by the member of the group has been affected by the pandemic. 5 Paragraphs 1.59-1.60 of Chapter I of the OECD TPG.

The unprecedented change in the economic environment following the outbreak of COVID-19 creates unique challenges for performing comparability analysis. The pandemic may have a significant impact on the pricing of some transactions between independent enterprises and may reduce the reliance that can be placed on historical data when performing comparability analyses. This may require taxpayers and tax administrations to consider practical approaches that can be adopted to address information deficiencies, such as comparability adjustments. Such practical approaches regarding the performance of comparability analyses should be consistent with the transfer pricing policy of the taxpayer over time.

The challenges associated with performing a comparability analysis may vary depending on the impact of the COVID-19 pandemic on the economically relevant characteristics of the accurately delineated transaction. For example, if a controlled transaction is covered by a pre-existing intercompany agreement (for example, if in 2018 it was determined that at arm’s length a party should receive an agreed fixed return for five years, and that parties at arm’s length would remain bound by that agreement), there may be no need to perform a comparability analysis for Financial Year (“FY”) 2020 provided that the facts and circumstances of the accurately delineated controlled transaction have not changed. In reaching this conclusion, it is important to consider any changes in the economically relevant characteristics, including the terms and conditions of the agreement, and whether at arm’s length, unrelated parties would have tried to renegotiate those terms and conditions.6 In contrast, where the arm’s length price of a controlled transaction is determined on an annual basis, it will be necessary to perform a comparability analysis for FY 2020. 6 See also the discussion in paragraphs 42 to 46 in this regard.

In principle, any form of publicly available information regarding the effect of COVID-19 on the business, industry and controlled transaction may be relevant in ascertaining the arm’s length nature of an enterprise’s transfer pricing policy implemented for FY 2020. The following sources of information may support that determination through the comparability analysis, generally by estimating the effect of the COVID-19 pandemic on the controlled transactions under review: An analysis of how sales volumes have changed during COVID-19, including whether the change is due to the use of other sales channels, and specifically compared to sales generated in pre-COVID years; An analysis of the change in capacity utilisation relevant for the MNE group and the controlled transaction7, and/or transactions with independent parties; Specific information relative to incremental or exceptional costs borne by parties to the controlled transaction (either with associated or unrelated parties) or by the MNE group as a whole; The extent to which government assistance has been received and, if so, quantifying the effect and identifying the type of the assistance and its accounting treatment; Details regarding government interventions that have affected the pricing and performance of controlled transactions; Information from interim financial statements such as quarterly SEC filings or earnings releases; Macroeconomic information like country specific GDP data or industry indicators from central banks, government agencies, industry or trade associations to the extent useful in understanding the context of the controlled transaction;8 Statistical methods such as regression analysis or variance analysis that are used to predict the extent to which a certain variable will vary with reference to other variables under certain specific conditions (e.g. the response of corporate profits in certain industries to GDP movements); A comparison of internal budgeted/forecasted data relating to sales, costs and profitability, compared to actual results; and An analysis of the effects on profitability or on third party behaviour observed in previous recessionary periods or using any data available in the current year, even if partial. 7 Paragraph 2.76 of Chapter II and illustration 3 in Annex I to Chapter II of the OECD TPG. 8 Paragraph 3.7 of Chapter III of the OECD TPG.

Another potential approach to utilise in setting transfer prices is to compare budgeted or forecast financial results to those actually achieved, to approximate the specific effects of COVID-19 on revenues, costs and margins. The financial outcomes that taxpayers within a controlled transaction would have achieved ‘but for’ the impact of COVID-19 may provide useful information, particularly when assessing the financial impacts from COVID-19 (e.g. reduced sales volume or increased operating expenses) and determining, in light of contractual terms and risk assumption of the parties, any appropriate resulting impact on intercompany prices. This analysis may include: The preparation of a detailed profit and loss analysis showing changes in revenue and expenses, with an explanation for variances resulting from COVID-19 – this may include a variance analysis of budgeted (pre-COVID) versus actual results; Details of profitability, adjusted to where the outcome would have been if COVID-19 had not occurred – this should consider all factors that have a positive or negative impact on the profits of the taxpayer to a controlled transaction and should be supported by evidence; The rationale and evidence for any increased allocation of costs or a reduction of sales (and subsequent changes in operating margins) to the tested party in the controlled transaction, taking into consideration its function, asset and risk profile; and Any evidence of any government assistance provided or affecting the tested party in the controlled transaction, its effect and its accounting treatment.

Such a review may be performed as part of a more general set of approaches utilised to evaluate the context and the factors that may impact the arm’s length nature of prices.

Information relating to the conditions of comparable uncontrolled transactions undertaken during the same period as the controlled transaction (“contemporaneous uncontrolled transactions”) is the most reliable information to use in a comparability analysis. Such information reflects how independent parties behave in an economic environment that is the same as or substantially similar to the economic environment of the controlled transaction.9 9 Section B of Chapter III of the OECD TPG.

In some instances, comparability analysis can be performed using contemporaneous (or near contemporaneous) uncontrolled transactions. For example, publicly available commercial databases typically have current or recent information on financial transactions between unrelated parties, which may provide reliable information on which to base comparability analyses under current economic conditions. Similarly, taxpayers are more likely to have current information on potential internal comparables, where these can be used to price related party transactions.

In other instances, it may be more challenging to use contemporaneous uncontrolled transactions as part of a comparability analysis, notably in the application of the transactional net margin method (“TNMM”). When applying the TNMM, taxpayers and tax administrations typically rely on historical information from commercial databases in order to set and test prices. FY 2020 information will typically not be available until mid FY 2021 at the earliest because commercial databases use publicly available information derived from financial statements and these financial statements tend to be lodged only after several months after the period to which they relate. This suggests that in these circumstances taxpayers will need to perform a comparability analysis based on available prior year financial information and, depending on the facts and circumstances of the case, utilising whatever current year information is available to support their transfer prices.

However, not every application of the TNMM will in principle require contemporaneous information for FY 2020. For example, a long term arrangement covering FY 2019 through FY 2022 may be in place, including an arm’s length price based on comparables contemporaneous with the negotiation of the arrangement, that insulates the tested party from risks that the tested party does not assume like those that play out during the pandemic. See also paragraph 10.

As the economic circumstances caused by the pandemic are continuing and evolving over time, taxpayers may encounter difficulties in determining arm’s length conditions due to the lag in time between the occurrence of controlled transactions and the availability of information regarding contemporaneous uncontrolled transactions.

Data from independent comparable transactions or companies from other time periods, such as average returns in preceding years, may not provide a sufficiently reliable benchmark for the current period without considering the specific impact of the pandemic on the controlled transactions under review.

The discussion below provides several pragmatic approaches to this issue. Tax administrations could consider these pragmatic approaches in an attempt to minimise disputes where taxpayers are making good faith efforts to determine arm’s length prices in the context of the information deficiencies associated with the COVID-19 pandemic. However, these approaches would not be appropriate in cases where taxpayers seek to use the circumstances attached to the COVID-19 pandemic to manipulate their pricing strategies in a way that is inconsistent with the arm’s length principle.

The difficulty posed by the delayed availability of contemporaneous data on comparable companies or transactions may have been exacerbated by the COVID-19 pandemic. Taxpayers and tax administrations should be mindful that determining a reliable arm’s length outcome requires flexibility and the exercise of good judgment.10 Difficult transfer pricing issues that arise as a result of the COVID-19 pandemic could give rise to a large number of mutual agreement procedure (“MAP”) disputes that could severely strain the resources of tax administrations. As such, tax administrations are encouraged to keep these complexities in mind when performing risk assessments, evaluating transfer pricing positions on audits and considering the support and documentation taxpayers provide that might demonstrate reasonable efforts and care when trying to comply with the arm’s length principle. Taxpayers should undertake reasonable and appropriate due diligence in evaluating the likely effects of the COVID-19 pandemic and in implementing appropriate changes in their transfer prices. MNE groups should document the best available market evidence currently available, which may be in the form of internal comparables, external comparables, or other relevant evidence of the economic impact of the COVID-19 pandemic (see paragraph 11), including its effects on the level of demand for goods and services, and on production and supply chains in particular sectors of the economy. 10 Paragraphs 1.13 and 2.74 of Chapter I and Chapter II of the OECD TPG.

The OECD TPG describe two approaches to identify and collect data required to undertake a transfer pricing analysis. The first is a “price-setting,” i.e. an ex-ante approach, which uses historical data updated to reflect any change in economic conditions through the date of the contract. The second is an “outcome-testing” approach, which may incorporate information that becomes available after the close of the taxable year to determine arm’s length conditions and report results on the taxable year to determine arm’s length conditions and report results on the tax return. According to the OECD TPG, both approaches, or a combination of these approaches, are found among OECD member countries.11 11 Paragraph 3.71 of Chapter III of the OECD TPG.

Where possible, and on a temporary basis during the pandemic, tax authorities that otherwise use the price-setting approach could consider allowing taxpayers, for those controlled transactions affected by the pandemic, to take into account information that becomes available after the close of the taxable year in filing their returns (where legally permissible and properly described in the transfer pricing documentation). Tax administrations could provide flexibility to allow amendments to FY 2020 tax returns such that transfer prices are set on an arm’s length basis and using available information. Also given the potential for double taxation that may arise as a result of unilateral adjustments, consideration may be given by tax administrations to: Provide for flexibility in the allowance of “compensating adjustments” to be made before the tax return is filed, where it is legally permissible, in order to allow for any available contemporaneous information to be better evaluated by taxpayers and tax administrations such that arm’s length prices can be reliably established12; or Ensure access to the MAP, or to some alternative applicable procedure, where the issue could be addressed between the respective tax administrations and early certainty could be obtained, to avoid double taxation, noting that through MAP or alternative procedures tax administrations can address issues in a non-adversarial proceeding, often achieving a negotiated settlement in the interests of all parties. 12 Paragraphs 4.38 and 4.39 of Chapter IV of the OECD TPG.

In the specific circumstances of the COVID-19 pandemic, the application of more than one transfer pricing method may be useful to corroborate the arm’s length price of a controlled transaction. In this context, it is important to note that the arm’s length principle does not require the application of more than one method and that the use of more than one method should follow the guidance in paragraphs 2.2 and 2.12 of the OECD TPG.

A comparability analysis that is solely based on financial information from the global financial crisis 2008/2009 would raise significant concerns (despite the obvious superficial similarities between that crisis and the current pandemic), given the unique and unprecedented nature of the COVID-19 pandemic and its effect on economic conditions, as well as the variability of the impact by business sector of the 2008/2009 crisis. In all cases, a comparability analysis should be performed by reference to the specific delineation of the controlled transaction, including its actual economic circumstances.

The principles outlined in Section B.5 of Chapter III of the OECD TPG regarding the use of multiple year data and averages remain applicable. In ordinary circumstances, the use of multiple year data and multiple year averages for comparability analyses may have certain advantages. For example, it can be used as a means to mitigate the impact of accounting differences, appropriately measure the effects on profitability for the tested party based on its business and product life cycles, and to evaluate the same for the comparables, such that the reliability of the comparison is increased.13 13 Paragraph 3.77 of Chapter III of the OECD TPG.

As a pragmatic means of addressing divergent economic conditions in the pre- or post-pandemic period, and when the pandemic was in effect and its effects on economic conditions were material, it may be appropriate to have separate testing periods (and periods considered for price setting) for the duration of the pandemic or for the period when certain material effects of the pandemic were most evident. This may be appropriate, so long as the data from independent comparables can be measured over a similar period in a consistent manner. Care should be taken to ensure the financial data of years affected by the pandemic do not unduly distort results from pre- or post-pandemic periods. In addition, government intervention in a market may materially affect the performance of activities. For example, in certain situations, the activities that otherwise normally would have occurred absent the pandemic may not occur in the same manner (or at all) during the period that the government intervention is in place. Also, in some cases a government intervention may permit activities to proceed that otherwise would have been curtailed or stopped by the pandemic. The accurate delineation of a controlled transaction will determine the effect, if any, of such intervention on the price or form of any controlled transaction associated with such activities.

This aspect is also relevant in performing the comparability analysis. For instance, assume government intervention forces a taxpayer to close its distribution facilities for three months. In undertaking a benchmark analysis, care should be taken in verifying that comparable enterprises have faced similar restrictions or conditions. Otherwise, it might be necessary to adjust the period over which the comparison is performed (e.g. excluding the economic data corresponding to the three months where the taxpayer was unable to operate). Taxpayers and tax administrations should determine on a case-by-case basis the extent to which these adjustments are necessary in circumstances where the potential differences may not have a material impact on the comparability. In this respect, the guidance in paragraphs 3.50 to 3.52 of the OECD TPG is relevant.

As with other analyses under the OECD TPG, numerous considerations may come into play, including the availability and choice of potential transfer pricing methods and comparables, and the interrelationship among them and the parameters of the testing periods (e.g. a transaction-based method may have a different time frame from a profit-based method). Just as it may improve reliability to use separate or more carefully circumscribed testing periods (or price setting periods) in some fact patterns (see paragraph 27), in other fact patterns the use of combined periods (that include both years that are impacted by the pandemic and years that are not impacted) may improve reliability.14 This approach would aggregate the financial results of FY2020, which may be exceptional, with the more normal results of prior years in order to test the arm’s length nature of the transfer pricing policy applied in FY2020. 14 Paragraphs 3.75 and 3.79 of Chapter III of the OECD TPG.

One potential solution to the uncertainty caused by the COVID-19 pandemic would be to allow for the inclusion of price adjustment mechanisms in controlled transactions. This may provide for flexibility while maintaining an arm’s length outcome. In particular, this approach to the extent permissible by domestic law would allow the adjustment of prices relevant for FY2020 through adjusted invoicing or intercompany payments effectuated in a later period (likely FY2021), when more accurate information to establish the arm’s length transfer price becomes available. In jurisdictions that use the outcome-testing approach, price adjustment mechanisms to reflect updated information relevant to determining an arm’s length price are often used. A jurisdiction that temporarily allows the outcome-testing approach could also temporarily allow the use of price adjustment mechanisms for that purpose and the taxpayer would be expected to describe the application of the price adjustment mechanism in its transfer pricing documentation. Such price adjustment mechanisms (provided that they are consistent with the arm’s length principle in the particular facts and circumstances) would address the issue of the lack of contemporaneous information on comparables or other direct evidence of arm’s length behaviour in response to the pandemic. This would give flexibility to taxpayers and tax administrations while also ensuring ultimate compliance with the arm’s length principle; however, given the scope of the potential adjustments, care would need to be taken with their appropriate characterisation, any effects that the payment may have on the comparability analysis for FY2021, and their potential resultant VAT/GST and customs duty implications (which are not the subject of this chapter or guidance).

The COVID-19 pandemic has created economic conditions that often differ from those of previous years. In these circumstances, where a taxpayer rolls forward an existing set of comparables to cover FY2020, it may be necessary to review the suitability of these existing comparables and potentially in some cases, it may be useful to revise the set, based on updated search criteria.

For example, assume that geographic comparability is deemed as the most relevant comparability factor given the nature of the effects of COVID-19 in a particular market. In these circumstances, in order to obtain reliable data from a particular market it may potentially be necessary to relax other comparability criteria, and then refine the sample.

In general, there is no overriding rule on the inclusion or exclusion of loss making comparables in the OECD TPG.15 Accordingly, loss-making comparables that satisfy the comparability criteria in a particular case should not be rejected on the sole basis that they suffer losses in periods affected by the COVID-19 pandemic.16 Consequently, when performing a comparability analysis for FY 2020, it may be appropriate to include loss-making comparables when the accurate delineation of the transaction indicates that those comparables are reliable (e.g. the comparables assume similar levels of risk and that have been similarly impacted by the pandemic). 15 Paragraph 3.64 of Chapter III of the OECD TPG. 16 Paragraph 3.65 of Chapter III of the OECD TPG.

During the COVID-19 pandemic, many MNE groups have incurred losses due to a decrease in demand, inability to obtain or supply products or services or as a result of exceptional, non-recurring operating costs.17 The allocation of losses between associated entities can give rise to dispute and hence is an issue that requires consideration given the probable increase in the frequency and magnitude of losses in the current economic environment. When considering the issue of losses and the allocation of COVID-19 specific costs, three issues warrant specific discussion. 17 For example, this might include expenditure on personal protective equipment, on IT infrastructure required to implement a “test and trace” system, measures to reconfigure office space to implement physical distancing requirements, or on other health-related safety equipment.

First, it is important to emphasise that the allocation of risks between the parties to an arrangement affects how profits or losses resulting from the transaction are allocated at arm’s length through the pricing of the transaction.18 Hence, the existing guidance on the analysis of risks in commercial or financial relations will be particularly relevant for determining how losses are allocated between associated parties. 18 Paragraph 1.58 of Chapter I of the OECD TPG.

Second, it will be necessary to consider how exceptional, non-recurring operating costs arising as a result of COVID-19 should be allocated between associated parties.19 These costs should be allocated based on an assessment of how independent enterprises under comparable circumstances operate. Separately, as extraordinary costs may be recognised as either operating or non-operating items, comparability adjustments may be necessary to improve the reliability of a comparability analysis. It is important to keep in mind that the treatment in a transfer pricing analysis of “exceptional,” “non-recurring,” or “extraordinary” costs incurred as a result of the pandemic will not be dictated by the label applied to such costs, but by an accurate delineation of the transaction, an analysis of the risks assumed by the parties to the intercompany transaction, an understanding of how independent enterprises may reflect such costs in arm’s length prices, and ultimately how such costs may impact prices charged in transactions between the associated enterprises (see OECD TPG paragraph 2.86, for example). Financial accounting standards should be considered in the comparability study, as they contain relevant and potentially helpful concepts in identifying the nature of costs. However, it should also be noted that even under those financial accounting concepts, there can be uncertainty as to whether particular costs are properly characterised as exceptional or extraordinary costs. 19 Depending on the duration of COVID-19 and the broader effects of the pandemic, the question may arise what constitutes an “exceptional, non-recurring” operating cost and when should such costs no longer be considered “exceptional” or “non-recurring”. As the effects of the pandemic vary by industry, business model or market, it is likely that this question can only be answered through a careful analysis of the specific costs under consideration.

Finally, the COVID-19 pandemic has created conditions in which associated parties may consider whether they have the option to apply force majeure clauses, revoke or otherwise revise their intercompany agreements. This may impact the allocation of losses and COVID-19 specific costs between associated parties, and therefore also requires specific consideration in the current economic environment.

When performing transfer pricing analyses, the activities performed by an entity may lead it to be characterised as “limited-risk” where it has a relatively lower level of functions and risks. 20 Though the term “limited-risk” is commonly used, since the term is not defined in the OECD TPG, the functions performed, assets used and risks assumed by “limited-risk” entities vary, and therefore it is not possible to establish a general rule that entities so-described should or should not incur losses. It should also be noted that neither the mere labelling of activities as “limited-risk” nor the fact that an entity receives a fixed remuneration means by itself that an entity operates on a limited risk basis in a controlled transaction.21 Further, no supposition should be made regarding the most appropriate transfer pricing method to apply in any set of circumstances without first undertaking a full and accurate delineation of the transaction, which then will help inform the choice of method made when performing the appropriate comparability analysis. 20 Paragraph 9.2 of Chapter IX of the OECD TPG. 21  Paragraph 1.81 of Chapter I of the OECD TPG.

In all circumstances it will be necessary to consider the specific facts and circumstances when determining whether a so-called “limited-risk” entity could incur losses at arm’s length. This is reflected in the OECD TPG which states that “simple or low risk functions in particular are not expected to generate losses for a long period of time”,22 and therefore holds open the possibility that simple or low risk functions may incur losses in the short-run. In particular, when examining the specific facts and circumstances, the analysis should be informed by the accurate delineation of the transaction and the performance of a robust comparability analysis. For example, where the losses incurred by third parties reflect a level of risks that is not comparable to the one assumed by the taxpayer in its controlled transaction then such a comparable should be excluded from the list of comparables (see paragraph 3.65 of the OECD TPG). 22  Paragraph 3.64 of Chapter III of the OECD TPG.

In determining whether or not a “limited-risk” entity may incur losses, the risks assumed by an entity will be particularly important. This reflects the fact that at arm’s length, the allocation of risks between the parties to an arrangement affects how profits or losses resulting from the transaction are allocated.23 For example, where there is a significant decline in demand due to COVID-19, a “limited-risk” distributor (classified as such, for example, based on limited inventory ownership – such as through the use of “flash title” and drop-shipping – and therefore limited risk of inventory obsolescence) that assumes some marketplace risk (based on the accurate delineation of the transaction) may at arm’s length earn a loss associated with the playing out of this risk. The extent of the loss that may be earned at arm’s length will be determined by the conditions and the economically relevant characteristics of the accurately delineated transaction compared to those of comparable uncontrolled transactions, including application of the most appropriate transfer pricing method and following the guidance in Chapter II of this note and Chapters II and III of the OECD TPG. In the example provided in this paragraph, the TNMM or potentially the resale- minus method depending on the more detailed facts and circumstances, might be used as the most appropriate method to test the arm’s length nature of the return, and third party comparable distributors might in these circumstances earn a loss, which may, for example, arise if the decline in demand means that the value of sales is insufficient to cover local fixed costs. It should be noted that the comparables chosen should be suitable in light of the accurate delineation of the transaction, in particular with reference to the risks assumed by each of the counterparties to the transaction. However, it will not be appropriate for a “limited-risk” distributor that does not assume any marketplace risk or another specific risk to bear a portion of the loss associated with the playing out of that risk. For instance, a “limited risk” distributor that does not assume credit risk should not bear losses derived from the playing out of the credit risk. For this reason, when determining whether an entity operating under limited risk arrangements can sustain losses the guidance in Chapter I of the OECD TPG, particularly as it relates to the analysis of risks in commercial or financial relations,24 will be particularly relevant. 23 Paragraph 1.58 of Chapter I of the OECD TPG. 24  Paragraphs 1.56 -1.106 of Chapter I of the OECD TPG.

When considering the risks assumed by a party to a controlled transaction, tax administrations should carefully consider the commercial rationale for any purported change in the risks assumed by a party before and after the outbreak of COVID-19 (and taking into consideration the accurate delineation of such purported change). In particular, concerns may arise where before the outbreak of COVID-19 a taxpayer argues that a “limited-risk” distributor did not assume any marketplace risk and hence was only entitled to a low return, but after the outbreak argues that the same distributor assumes some marketplace risk (for example, due to changes in risk management functions) and hence should be allocated In this scenario, consideration should be given to re-examining whether prior to the outbreak of COVID-19 the “limited-risk” distributor genuinely did not assume any marketplace risk, whether after the outbreak the “limited risk” distributor did not actually assume any marketplace risk, and/or whether the assumption of this risk following the outbreak of COVID-19 is a result of a business restructuring. If a prior risk allocation is recognised under an accurate delineation, in order for a reallocation of that risk to be recognised under a subsequent updated accurate delineation, such new risk allocation must be supported by an analysis of all the facts and circumstances and relevant evidence should be obtained and documented to substantiate the position. In this respect, the guidance in Chapter IX of the OECD TPG may be relevant. In general, consideration should be given to whether a taxpayer is taking inconsistent positions pre- and post- pandemic and, if so, whether either position is consistent with the accurate delineation of the transaction.

In response to the COVID-19 pandemic, independent parties could seek to renegotiate certain terms in their existing agreements.25 Associated parties may also consider revising their intercompany agreements and/or their conduct in their commercial relationships. Tax administrations should therefore review the agreements and/or the conduct of associated enterprises, in light of the guidance in section D of Chapter I of the OECD TPG, together with observations of relevant behaviour of independent parties and this guidance, in order to ascertain whether any such renegotiation should be respected under the OECD TPG. The accurate delineation of the controlled transaction will determine whether the revision of intercompany agreements is consistent with the behaviour of unrelated parties operating under comparable circumstances. 25  Part 1, Section F of Chapter IX of the OECD TPG.

Given the current economic environment, it is possible that independent parties may not strictly hold another party to their contractual obligations, particularly if it is in the interest of both parties to renegotiate the contract or to amend certain aspects of their For example, unrelated enterprises may opt to renegotiate a contract to support the financial survival of any of the transactional counterparties given the potential costs or business disruptions of enforcing the contractual obligations, or in view of anticipated increased future business with the counterparty. This behaviour should be considered when determining whether or not associated parties would agree to revise their intercompany agreements in response to COVID-19.

For example, assume that Distributor X purchases products, the controlled transaction, from a related party Company Y, and sells these products to third party customers. Further assume that a major customer of Distributor X does not pay for products purchased within its standard 30-day term, and that this causes a cash flow issue for Distributor X, who bears credit risk under the accurately delineated transaction. Under these circumstances, Distributor X may seek to renegotiate its payment terms on a temporary basis with Company Y. The determination of whether this renegotiation is arm’s length should be based on what independent parties would do under comparable circumstances and if there have been situations at arm’s length where contractual terms have not been enforced, or have been amended, this may form reasonable evidence for taxpayers to justify revised terms in intra-group agreements where the situations are comparable.

Determining whether a renegotiation of a commercial arrangement (including pricing under the arrangement going forward and any potential compensation for the renegotiation itself) represents the best interests of the parties to a transaction requires careful consideration of their options realistically available26 and the long-run effects on the profit potential of the parties.27 For example, an entity may agree to restructure a transaction if the alternative option is losing a key customer or supplier, where it considers that the restructuring will maximise its profits in the long-run. Consideration should also be given to whether the economic impact resulting from the renegotiation may require indemnification (as defined in OECD TPG paragraph 9.75) of the harmed party.28 26 It should be noted that in an uncontrolled transaction one party might attempt to force a renegotiation by threatening to violate the terms of an existing agreement, believing that the other party will not find it worthwhile to seek judicial enforcement of the agreement, whereas this course of action may not realistically be available in the context of a controlled transaction. 27 Paragraphs 9.78-9.97 of Chapter IX of the OECD TPG. 28 Paragraphs 9.78-9.97 of Chapter IX of the OECD TPG.

The above analysis outlines the factors that should be considered when determining whether associated parties may at arm’s length consider revising their intercompany agreements and/or their conduct in their commercial relationships as a consequence of the COVID-19 pandemic. However, it is important to emphasise that in the absence of clear evidence that independent parties in comparable circumstances would have revised their existing agreements or commercial relations, the modification of existing intercompany arrangements and/or the commercial relationships of associated parties is not consistent with the arm’s length principle. Accordingly, such modifications should be treated with caution and well-supported by documentation outlining how the modification is in line with the arm’s length principle.

As a result of the COVID-19 pandemic, many enterprises have incurred exceptional, non-recurring operating costs relevant to differing operating conditions for the pandemic period. These include expenditure on Personal Protective Equipment (PPE), reconfiguration of workspaces to enable physical distancing, IT infrastructure expenses relating to test, track and trace obligations and to implement teleworking arrangements. In determining how these costs should be allocated between related parties, it will be important to consider how these costs would be allocated between independent parties operating in comparable circumstances.

Allocation of operating or exceptional costs would follow risk assumption and how third parties would treat such costs. Thus in order to determine which associated enterprise should bear exceptional costs, it would be first necessary to accurately delineate the controlled transaction, which would indicate who has the responsibility for performing activities related to the relevant costs and who assumes risks related to these activities. For example, if a cost directly relates to a particular risk, then the party assuming that risk would typically bear the costs associated with that risk. Furthermore, the party initially incurring an exceptional cost may not be the party assuming risks associated to that cost at arm’s length, and consequently such costs may need to be passed on to parties that do assume such risks. Thus a thorough analysis should be performed before concluding whether all or part of the operating or exceptional costs should be allocated between related parties.

Further, it should be noted that certain operating costs may not be viewed as exceptional or non- recurring in circumstances where the costs relate to long-term or permanent changes in the manner in which businesses operate. For example, certain costs relating to teleworking arrangements may become permanent if working from home became more common as a result of the pandemic. Consequently, if the expense is viewed as neither being exceptional nor non-recurring and reflects more common means of doing business, then it should be treated as such when delineating the transaction to which the costs pertain and in undertaking the comparability analysis. Furthermore, it should also be noted that for certain businesses the COVID-19 pandemic has led to reduction in or elimination of certain costs that were typically incurred prior to the COVID-19 pandemic. These will differ depending on the underlying facts and circumstances, but might include expenses on rent, the day-to-day running expenses of a physical office, and travel related expenses, among others. Expenses that relate to substitutes to the means of conducting business activities would likely be treated as operating costs, depending on the underlying facts and circumstances. If teleworking costs, or any other costs pertaining to remote working facilities, are centrally borne by one entity of the MNE group, it may be appropriate to charge out such expenses to parties that benefit from the underlying product or service to which the expense relates.

At arm’s length, exceptional costs may or may not be passed on (wholly or partially) to customers or suppliers depending on who has the responsibility to bear such costs and (including in cases in which such responsibility is not expressly provided for) the consequences of the accurate delineation of the controlled transaction (including risk assumption) and the comparability analysis. For example, which party ultimately bears such costs might be influenced by the competitiveness of the industry within which the activity occurs and how demand responds to changes in price. For example, a manufacturer in a highly competitive market, with undifferentiated products, may be unable to pass on exceptional costs to its customers, without experiencing a decline in demand for its services (unless its competitors are passing on similar costs). However, a similar manufacturer that produces differentiated products in a comparatively uncompetitive industry may be able to pass on these costs to its customers, at least partially, without experiencing a decline in demand.

When performing a comparability analysis, it may be necessary to specifically consider how exceptional costs arising from COVID-19 should be taken into account.

First, exceptional costs should generally be excluded from the net profit indicator except when those costs relate to the controlled transaction as accurately delineated.29 The exclusion of exceptional costs must be done consistently at the level of the tested party and the comparables to ensure a reliable outcome, noting that the availability of this information may be limited.30 Care should be taken in order to ensure that such costs are appropriately measured and are consistently accounted for to the extent possible. 29 Paragraph 2.86 of Chapter II of the OECD TPG.30 Paragraph 2.74 of Chapter II of the OECD TPG.

Second, when determining a cost basis, it will be important to consider whether the basis should include or exclude exceptional costs that are deemed to relate to the controlled transactions (determination noted above), and, if included in the costs basis, whether such costs should or should not be treated as pass-through costs to which no profit element should be attributed (see paragraph 2.99 of the OECD TPG). Including exceptional costs in the cost basis would transfer these costs to the counterparty, whereas excluding them would have the effect of allocating them to the tested party. Therefore, in determining which approach is most appropriate, it will be important to consider at arm’s length which party to the controlled transaction would have borne these additional costs, which should in turn be informed by the accurate delineation of the transaction.31 31 Paragraph 2.51 and 2.98 of Chapter II of the OECD TPG.

Third, adjustments for accounting consistency may be required to improve comparability. Adjustments for accounting consistency are designed to eliminate the effect of differing accounting practices between the controlled and uncontrolled transactions and should be considered if and only if they are expected to increase the reliability of the results of a comparability analysis.32 In some cases, if exceptional costs arising from COVID-19 may be accounted for as either operating or non-operating items by different taxpayers in different transactions, then comparability adjustments may be In other cases there can be differences in whether the COVID-19 related costs are taken into account above or below the gross profit line. For instance, the recognition of the purchase of PPE as an operating cost by the tested party and as a cost of goods sold by a comparable may have a significant impact when computing a profit level indicator based on gross profit and may require a comparability adjustment. 32 Paragraph 3.48 and 3.50 of Chapter III of the OECD TPG.

Force majeure clauses may be invoked in order to suspend, defer, or release an enterprise from its contractual duties without liability in certain situations.33 This may result in losses for enterprises because of the loss of a customer, supplier or an ordinarily profitable contract, and could also lead to the closure of business operations and associated restructuring costs. 33 Note that these guidelines do not seek to legally define concepts such as “force majeure” or provide comment on when it may legally be invoked, but instead focus on the transfer pricing implications of the existence of the force majeure concept and its invocation. The “force majeure” concept originated in civil law systems. While the doctrine does not apply automatically in all civil law countries, certain European civil law countries at least implicitly recognise the force majeure principle in their civil codes (i.e. it may not be necessary to include it in a contract because the statutory force majeure provisions apply automatically to all contracts within those jurisdictions). In common law jurisdictions, the application of force majeure is not implied in contracts; thus, parties to the contract need to include that clause expressly, detailing the specific circumstances under which the parties can suspend or discontinue performance of its contractual obligations. In common law jurisdictions, in the absence of a “force majeure” clause in the contract, the parties may invoke certain common law doctrines to attempt to end and release the parties’ contractual obligations.

Because of the COVID-19 pandemic, a party may attempt to assert that the extreme circumstances justify the non-performance of a contract and this may be achieved through invoking a force majeure clause, which defines circumstances beyond the control of parties to a transaction that can frustrate or render impossible contractual performance. For instance, force majeure events arising in the context of COVID-19 could be the prohibition of activities by a governmental body, for example through the enforced closure of production or retail facilities.

Where one party to a controlled transaction seeks to invoke force majeure, the agreement and underlying legal framework within which force majeure may be invoked should form the starting point of a transfer pricing analysis. It cannot be automatically assumed that where a relevant intercompany contract contains a force majeure clause that the COVID-19 pandemic is sufficient for a party to that contract to invoke force majeure, nor can it be automatically assumed in the absence of such a clause in the intercompany contract that a renegotiation with a potentially similar outcome at arm’s length would be inappropriate (see paragraph 59 below). Whether COVID-19 constitutes a force majeure in a particular case will depend on the plain language of the force majeure provision (and possibly also on how that provision interacts with other terms, such as certain terms of the controlled transaction itself). In addition, it will be relevant to analyse the conduct of the parties in reviewing an existing force majeure provision or in ascertaining whether or not it may be asserted in the absence of a specific term. The accurate delineation of the controlled transaction will determine whether invoking force majeure is permissible, including by reference to the conduct of the parties and not just by reference to the legal agreement. Care should be taken to assess whether the magnitude of the disruption caused by COVID-19 in the specific related party situation qualifies as a force majeure event, and to review the force majeure clause in the context of the overall relationship and contractual agreement. An analysis of the economic circumstances of the commercial arrangement is relevant to determining whether, at arm’s length, a party would decide to invoke a force majeure clause.

For example, assume that Company G in Jurisdiction G provides manufacturing services to Company H under a long-term manufacturing services agreement that includes a force majeure clause. The government in jurisdiction G mandates the closure of the manufacturing facility for a certain specified short-term period, which may be extended depending on the duration of the pandemic. Given the lack of clarity on the extent of the disruption, it would be important to analyse the contract to see if the disruption qualifies as a force majeure event and consider whether, at arm’s length, Company G or Company H would seek to invoke the clause. Assuming that a clause may be legally invoked under the relevant legal framework, given the long-term nature of the relationship and the short-term nature of the disruption, it may be the case that neither company would invoke the clause, even if it did qualify as a force majeure event. If the disruption was for a longer period, then the circumstances may be different, and force majeure may be more likely invoked.

In response to COVID-19, some taxpayers may seek to assert force majeure in situations where it is not contained within the relevant intercompany agreement (assuming here that the law governing the contract is not a civil law jurisdiction where force majeure would automatically apply), may seek to change an existing intercompany agreement to insert a force majeure clause, or may seek to assert that a renegotiation at arm’s length would have similar economic outcomes. In these circumstances, tax administrations should carefully review such assertions in light of the accurately delineated transaction (including consideration of the conduct of the parties, both past and present) and the economically relevant circumstances of the transaction. Tax administrations should therefore review the agreements and/or the conduct of associated enterprises, in light of the guidance in section D of Chapter I of the OECD TPG, together with observations of relevant behaviour of independent parties and this guidance, in order to ascertain whether any such assertion, revision or renegotiation should be respected under the OECD TPG, and that the transfer pricing outcomes are appropriate in light of the accurate delineation of the transaction.

Government assistance is a monetary or non-monetary programme where a government or other public authority provides a direct or indirect economic benefit to eligible taxpayers such as grants, subsidies, forgivable loans, tax deductions, or investment allowances. For example, a government may directly subsidise the labour costs incurred in undertaking certain activities or may indirectly support businesses through the provision of local infrastructure, such as a business park.

During the COVID-19 pandemic, governments’ overriding concern has been public health and controlling the spread of the virus while at the same time trying to help enterprises manage the impact of the decline of business activity and workers facing a decline in employment opportunities and income.

Job retention programmes have been used in many jurisdictions to preserve jobs in enterprises experiencing a temporary reduction in business activity.34 Examples of these programmes include short- term work programmes that directly subsidise hours not worked35 and wage subsidy programmes that subsidise hours worked or top up the earnings of workers on reduced hours.36 34 A crucial aspect of all job retention schemes is that employees keep their contracts with the employer even if their work is suspended. 35 For example, the German Kurzarbeit or the French Activité partielle. 36 For example, the Dutch Emergency Bridging Measure (Noodmatregel Overbrugging Werkgelegenheid, or “NOW”) or the Job Keeper Payment in Australia.

Governments have also provided broader financial and liquidity supports to ensure enterprises can continue to operate through the period of reduction in business activity. Those include: (i) loan guarantees; (ii) direct financing to business on preferential terms; (iii) loan deferrals; (iv) specific grants and (v) tax relief.

The availability, substance, duration and take-up of these programmes potentially have transfer pricing implications, whether the government assistance is provided to a member of an MNE group directly or made available to independent parties within the market where an MNE group operates (thus affecting the behaviour of enterprises engaged in potentially comparable transactions).37 37 This chapter deals with government assistance programmes. This chapter does not discuss the effects of other government interventions, including interventions that prevent or limit a party’s ability to fulfil an existing intercompany agreement.

The terms and conditions of government assistance programmes related to COVID-19 need to be considered when determining the potential impact of these programmes on controlled transactions and when comparing their effects with those of other pre-existing assistance programmes. For instance, a number of COVID-19 assistance programmes are designed as temporary support to preserve businesses as a going concern and their impact in the transfer pricing analysis may differ from the impact of ongoing assistance programmes (linked or not to COVID-19 assistance), the duration of which may cover several years.

In addition, there may be challenges in establishing the nature of government assistance received by potential comparables, given the different types of COVID-19 government assistance programmes, the practical difficulties to obtain detailed and reliable information and the delay in data availability. In this context, the analysis required on the specific characteristics of the government assistance should take into account the economic impact of the assistance on the accurately delineated transaction. Therefore, an exhaustive analysis of the specific characteristics of government assistance would not be required in circumstances where the receipt of government assistance is unlikely to have a material impact on the accurately delineated controlled transaction. (See section C of Chapter III of the OECD TPG)

Economically relevant characteristics that pertain to a controlled transaction should be evaluated: (i) when accurately delineating the controlled transaction; and (ii) to facilitate the comparison between the accurately delineated controlled transaction and comparable uncontrolled transactions, in order to price the controlled transaction.38 38 Paragraph 1.33 of Chapter I of the OECD TPG.

The extent to which the receipt of government assistance is an economically relevant characteristic may vary. An example of where government assistance may be more economically relevant is the provision of a wage subsidy, a government debt guarantee or short-term liquidity support. In such circumstances, the receipt of government assistance may have a direct impact on the controlled transaction and comparable transactions between independent parties, including their prices. In other situations the receipt of government assistance may be less economically relevant. For example, the provision of local infrastructure by a government might be only indirectly linked to the controlled transaction and the compensation thereof. Additionally, there may be other situations where the parties to a controlled transaction do not receive government assistance, but another party does, and this may influence the economically relevant characteristics of the transaction.

The determination of the economic relevance of government support will inform its effect, if any, on accurately delineating the controlled transaction and performing the comparability analysis. If the government assistance is an economically relevant characteristic, this information should be included as a part of the documentation to support the transfer pricing analysis.

Section D.4 of Chapter I of the OECD TPG elaborates on the effect of government policies noting that, as a general rule, government interventions should be treated as conditions of the market in the particular country. Therefore, the receipt of government assistance may be part of the economic circumstances of the parties and a feature of the market in which the parties operate. In this regard, the provisions in section D.6.2 of Chapter I of the OECD TPG on other local market features may provide guidance relevant to the transfer pricing implications of government assistance either directly or by analogy.

In particular, by analogy to section D.6.2 of Chapter I of the OECD TPG, the analysis of the implications of the receipt of government assistance would need to consider the following factors (not only to identify reliable market comparables when they exist, but also in cases where such comparables cannot be identified): whether the receipt of government assistance provides a market advantage to the recipient; the amount of any increase in revenues, decrease in costs, vis-à-vis those of reliable comparables, that are attributable to the government assistance received, and the duration of the assistance; the degree to which benefits of government assistance, at arm’s length, are passed on to independent customers or suppliers (see paragraph 76); and, where benefits attributable to government assistance exist and are not fully passed on to independent customers or suppliers, the manner in which independent enterprises operating under similar circumstances would allocate such benefits between them.

The analysis of the receipt of government assistance as a local market feature may help to inform an analysis of whether its receipt affects the price of a controlled transaction. It may also be relevant when conducting a comparability analysis. However, the particular nature of the government assistance should always be reviewed in the transfer pricing analysis, as under certain circumstances it may not be a general feature of the market, and/or it may not generate benefits for either party.

The potential effect of the receipt of government assistance on the pricing of a controlled transaction will depend on the economically relevant characteristics of the transaction, following an accurate delineation of the controlled transaction and the performance of a comparability analysis. Therefore, it would be contrary to the arm’s length principle to assume that the mere receipt of government assistance would affect the price of the accurately delineated controlled transaction, without performing a careful comparability analysis (including an analysis of how the receipt of government assistance would affect the price of uncontrolled transactions, if at all, and the perspectives of both parties to the transaction).

The economically relevant characteristics of the accurately delineated controlled transaction will help in determining the potential effect of the receipt of government assistance on the pricing of the controlled transaction, if any. For instance, some of the aspects to consider in analysing the impact, if any, of the receipt of government assistance on the price of a controlled transaction include the availability, purpose, duration and other conditions imposed by the government in granting the assistance; the allocation of the economically significant risks; and the level of competition and demand within the relevant markets. In addition, as indicated in paragraph 1.34 of the OECD TPG, the identification of these economically relevant characteristics would require a broad evaluation of how the MNE group responds to the receipt of government assistance.

Government assistance programmes may be subject to a number of legal conditions that could limit or even prevent the capacity of the party receiving the assistance from modifying the pricing of its transactions with other parties across the value chain and that should be taken into account in performing a comparability analysis, along with the other comparability factors described in paragraph 74. Similarly, government assistance programmes could be, for example, contingent upon eligibility criteria that could require the recipients to demonstrate a significant downturn in revenue or could be intended to preserve the viability of businesses during the pandemic. In addition, some government assistance programmes may be subject to uncertainty around the amounts involved. These aspects could limit the effect of government assistance on the price of the goods or services offered by the entity receiving the assistance and should be considered as part of the comparability analysis.

The economic circumstances of the market in which the party receiving government assistance operates could also influence the pricing of the accurately delineated controlled transaction. Subject to the specific facts and circumstances, aspects such as the level of competition, the elasticity of the demand or the availability of the benefits to competitors in the market, which may not be known or disclosed, could be relevant factors in determining whether the receipt of government assistance would be translated into an entity’s pricing strategy. In addition, the economic circumstances of the market in which the MNE group sells its products to third party customers would also be relevant to determine, for instance, whether the receipt of government assistance has led the MNE group to change its pricing strategies towards unrelated customers, either to retain the assistance within the MNE group or to pass it on to third parties.

The analysis of the effect of government assistance on the price of a controlled transaction will take into account the allocation of the economically significant risks under the accurate delineation of the controlled transaction, the impact of the pandemic on the outcome of the economically significant risks, and the linkage between the type of government assistance and those risks. Based on the facts and circumstances, the impact of government assistance on the price of a controlled transaction, if any, would depend, among other things, on which party assumes the economically significant risks affected by the pandemic in the context of an analysis of the accurately delineated controlled transaction. In order to reach a conclusion, consideration also should be given to all the other comparability factors described in paragraph 74.

Under the guidance in Chapter II of the OECD TPG, when establishing arm’s length prices using one-sided methods, particular care must be taken to avoid adopting without further analysis a particular mechanical approach (such as offsetting cost savings achieved through government assistance against the relevant cost base for the transaction; recognising government assistance as revenue; or recognising government assistance as extraordinary income) since this could lead to non-arm’s length prices in transactions among associated parties.

In the absence of reliable comparables or other reliable information (such as a robust analysis of the economically relevant characteristics described in paragraph 74) regarding how independent parties would allocate government assistance, caution should be exercised in assessing whether a purported sharing of government assistance represents an arm’s length outcome.

The receipt of government assistance may reduce the quantitative negative impact of a risk. For instance, a party assuming credit risk could expect to incur losses from a transaction due to financial difficulties of its counterparty. However, the counterparty may in fact be able to meet its obligations by benefitting from government assistance. This aspect (i.e. the reduction in the negative impact of risk) must be distinguished from the allocation of risk under the guidance in section D.1.2.1 of Chapter I of the OECD TPG.

Under the guidance of Chapter I of the OECD TPG, the provision of government assistance to an associated party will not change the allocation of risk in a controlled transaction for transfer pricing purposes. For instance, assume Company W is a distributor that purchases goods from a related party manufacturer and sells those goods to third party customers in the jurisdiction where it is resident (Country W). Under the accurate delineation of the transaction, the marketplace risk is assumed by Company W. This implies that, under normal economic circumstances, including economic cycles, Company W bears the consequences of the playing out of the marketplace risk (e.g. a decrease in demand due to new competitors entering the market). Assume further that demand for Company W’s products declines significantly due to the measures adopted by the government in Country W in response to the COVID-19 pandemic. Company W receives government assistance in the form of a cash grant to help the entity to support its fixed and operating costs during the period where the measures remain in place. Under the guidance in paragraphs 1.66 and 1.67 of Chapter I of the OECD TPG, the receipt of government assistance does not alter Company W’s contractual arrangements and does not alter Company W’s capability and actual performance of decision-making functions relating to the marketplace risk. Despite the government intervention, Company W would retain the competence and experience regarding the marketplace risk and would continue to possess an understanding of the impact of its decisions related to that risk on the business. Therefore, under the prevailing facts and circumstances, the government assistance to support Company W in tackling the financial distress derived from the COVID-19 pandemic does not modify the allocation of the marketplace risk to Company W. The same conclusion would apply to other risks that had been equally allocated to the distributor, e.g. inventory risk or credit risk.

The comparability of open market transactions or enterprises may be influenced by the receipt of government assistance, affecting both how the parties establish their commercial or financial relations and how they price their transactions. Therefore, when performing a comparability analysis, it may be necessary to take into account the receipt of government assistance when reviewing potential comparables.

For example, as government assistance and the specific circumstances of the COVID-19 pandemic may vary across different markets, it may affect the comparables and the arm’s length prices of uncontrolled transactions in different ways.39 For instance, assume Company D, a member of an MNE group, provides manufacturing services and receives a domestic wage subsidy under a job retention programme offered by the government of Country D. Assume further that third party manufacturer Company E in Country E receives a domestic short-time work programme where the hours not worked by employees are directly subsidised. The services provided by Company D and Company E may be similar but the general availability and receipt of government assistance to Company D and Company E may result in a material difference in the arm’s length conditions of the transactions for the period over which the short-time work programme is in place and may not be sufficient to ensure a reliable comparison. In that case, the services provided by Company E would not be comparable for purposes of evaluating and determining the arm’s length price for Company D’s services based on the principles of Chapters II and III of the OECD TPG unless comparability adjustments to take such differences into account can reliably be made, based on an analysis of all of the relevant facts and circumstances of the Other characteristics of the government assistance that might affect comparability, depending on the circumstances, could include, for example, the duration of an assistance program and the relation between the assistance provided and the pandemic-related costs or the lost revenue. 39 Paragraph 1.112 of Chapter I of the OECD TPG on the relevance of differences across markets.

The most reliable approach in identifying reliable comparables will be to refer, where possible, to data regarding comparable uncontrolled transactions in the same or comparable geographic market between independent enterprises performing similar functions, assuming similar risks, and using similar assets.

The materiality of the change in economically relevant circumstances created by the impact of government assistance available in a market may impose additional challenges to the comparability analysis. It may for example render it more difficult to apply the comparability analysis through the utilisation and/or the application of a particular transfer pricing method and search for comparable transactions, on the basis that comparability differences may be exacerbated due to variations in government assistance between comparables or between jurisdictions.40 For example, an uncontrolled transaction that might otherwise have been considered comparable to a particular controlled transaction might be considered not comparable by virtue of the fact that one of the transactions is subject to government assistance while the other is not. A revised strategy, and potentially the use of a corroborating transfer pricing methodology,41 may need to be applied in such cases to take account of the differences in comparability. See Chapter II of this guidance. 40 As the guidance in paragraph 2.143 of the OECD TPG indicates, the lack of comparables alone is insufficient to warrant the use of a transactional profit split. 41 In considering the use of more than one method, the guidance in paragraph 2.12 of the OECD TPG should be followed in any case.

Finally, when applying a one-sided method such as the resale price method, the cost plus method, or the TNMM, the accounting treatment of the government assistance in both the tested party and any comparable may need to be specifically identified, especially when the tested party and the comparables apply different accounting standards. For example, the government assistance may be deducted from the costs under the relevant accounting standard, or it may be presented separately. In addition, the accounting treatment of government subsidies under different accounting standards may impact different levels of profitability (e.g. gross profit, operating profit, net profit, etc.) or might even be accounted for in the “other comprehensive income” statement, only being recycled into the “profit or loss statement” of the entity over time. Where accounting treatments of the same type of assistance differ between the tested party and the comparable, a comparability adjustment may be required. In addition, divergences in the accounting treatment of government assistance could point to a difference in the type of government support provided – e.g. the accounting treatment of a conditional loan differs from that of an outright grant. Such a difference could affect comparability and might be more difficult to adjust for than a simple accounting difference.

COVID-19 has led to material changes in economic conditions that were not anticipated when many APAs covering FY2020 and potentially future financial years affected by COVID-19 were agreed. Given this situation, it is important to determine to what extent, if any, the change in economic conditions affects the application of existing APAs. Taxpayers and tax administrations negotiating APAs that apply to FY2020 may also face questions about how the economic conditions arising from COVID-19 should be taken into account. For this reason, this guidance explains the possible impact of COVID-19 on existing unilateral, bilateral and multilateral APAs and APAs under negotiation.

One of the primary benefits of an APA is that it provides tax certainty to taxpayers and tax administrations by ensuring predictability in the treatment of international transactions for tax purposes. It is important to underline, however, that taxpayers and tax authorities are encouraged to take constructive and collaborative approaches in the APA process, which are conducive to the long-run success of an APA programme.

Some taxpayers may face challenges applying existing APAs under the economic conditions resulting from the COVID-19 pandemic. In those instances, taxpayers are encouraged to adopt a collaborative and transparent approach by raising these issues with the relevant tax administrations in a timely manner. Taxpayers should not seek to resolve them unilaterally without consulting with the relevant tax administrations.

Yes, existing APAs and their terms should be respected, maintained and upheld, unless a condition leading to the cancellation or revision of the APA (e.g. breach of critical assumptions) has occurred. Taxpayers and tax administrations cannot automatically disregard or alter the terms of existing APAs due to the change in economic circumstances. 42 In considering how to respond to the impact of COVID-19 on existing APAs, the guidance provided, in Annex II to Chapter IV of the OECD TPG: Guidelines for Conducting Advance Pricing Arrangements under the Mutual Agreement Procedure (MAP APAs) is relevant.

Generally, the APA itself will explicitly describe what constitutes a situation of non-compliance or failure to meet a critical assumption, as well as the consequences arising from it. Also, domestic law or procedural provisions may also impose consequences or obligations on the taxpayer and affected tax administrations. All this should be considered by taxpayers and tax administrations in determining the impact of the COVID-19 pandemic on existing APAs.

Most APAs include critical assumptions about the operational and economic conditions that will affect the transactions covered by the APA. The COVID-19 pandemic and the response of governments have dramatically affected the economic and market conditions and are likely to qualify as a breach of the critical assumptions (para. 44 b) and c) of Annex II to Chapter IV of the OECD TPG). A mere change in business results during the period affected by the COVID-19 pandemic would not, however, result in a breach of a critical assumption (unless the particular APA had a critical assumption regarding changes in business results).

Whether there has been a breach in a critical assumption should be analysed on a case-by-case basis, and it should take into account the individual circumstances of the taxpayer and commercial environment. The COVID-19 pandemic has not had the same impact on all enterprises. While many industries and business have experienced a drop in demand and revenues because of forced lockdowns, others have expanded their consumer-base or benefitted from new business opportunities. Whether a breach has occurred may also depend on the duration of the disruption. If a breach has occurred, in determining an appropriate response, a tax administration should carefully consider the extent of the divergence between the agreed parameters in the APA and the new parameters under the COVID-19 economic circumstances43; and, the ability of the agreed transfer pricing methodology to reliably reflect arm’s length pricing of a controlled transaction under the new situation.44 43 Paragraph 45 of Annex II to Chapter IV of the OECD TPG. 44 Paragraph 43 of Annex II to Chapter IV of the OECD TPG.

Where tax administrations establish that the critical assumptions of an APA have not been breached, the existing APA, as agreed, must continue to be respected, maintained and upheld. If a taxpayer believes that the terms of the APA are no longer appropriate, it should not seek unilaterally to breach critical assumptions deliberately or fail to comply with the terms or conditions of the APA, and it should avoid making unreliable price adjustments or taking other actions that are not consistent with (or otherwise fail to comply with) the terms of the APA that it may view as not appropriate. Instead, where taxpayers have concerns, they should approach the relevant tax administration in a transparent way to discuss their concerns.

When considering the consequences of the failure to meet critical assumptions, tax administrations and taxpayers should consider the (i) terms of the APA; (ii) any agreement between relevant tax administrations as to how to deal with the failure; and (iii) any applicable domestic law or procedural provisions.45 This section follows the guidance in Chapter IV, Section F and Annex II to Chapter IV of the OECD TPG. 45 Paragraph 74 of Annex II to Chapter IV of the OECD TPG.

In some situations, the APA agreement, domestic law or procedural provisions may prescribe procedures to follow, or describe the consequences that will arise, in situations where there is a failure to fulfil critical In these situations, tax administrations should follow the prescribed procedures.46 In other situations, tax administrations may have some discretion over their response.47 Also in the event that the effect of the breach is not material, (Paragraph 75 of the Annex II to Chapter IV of the OECD TPG), the taxpayer and tax administration may agree to continue to apply the APA In the case of bilateral or multilateral APAs, a common and helpful practice is for the tax administrations to consult before imposing any unilateral changes,48 and such consultation is mandated in some APAs.49 In the absence of other rules and procedures prescribed by domestic law, when the guidance in Chapter IV of the OECD TPG apply , a breach of critical assumption with the APA could have three potential outcomes as provided in section E.3 Annex II to Chapter IV of OECD TPG:50 Revision, which means that the taxpayer and tax administrations still have the benefit of the APA for the whole of the proposed period, albeit that different terms apply before and after the revision date. Cancellation, which means the APA is treated as being effective and in force but only up to the cancellation date and not for the whole of the proposed period. Revocation, which has the effect that the taxpayer is treated as if the APA had never been entered into. 46 Paragraph 74 of Annex II to Chapter IV of the OECD TPG. 47Paragraph 66 of Annex II to Chapter IV of the OECD TPG, item e) (breach of critical assumption would “trigger renegotiation of the agreement”). 48 Paragraph 75 of Annex II to Chapter IV of the OECD TPG (considering what “the tax administrations determine”). 49 Paragraph 66 of Annex II to Chapter IV of the OECD TPG, item e) (breach of critical assumption would “trigger renegotiation of the agreement”). 50 Paragraph 76 of Annex II to Chapter IV of the OECD TPG.

Revision would be the appropriate response where there has been a material change in conditions noted in a critical assumption in the APA and the tax administration and the taxpayer agree on how to revise the APA.51 The revision of an existing APA would have the effect of changing the previously agreed terms of an APA. Under this approach, it is likely that the original terms of the APA would remain in force for the period up to FY2020, with revised terms applying for FY2020 and any subsequent year(s) subject to the economic circumstances derived from the COVID-19 pandemic. 51 Paragraphs 83 and 85 of Annex II to Chapter IV of the OECD TPG.

In some circumstances, it may be possible to retain some of the terms set out in an APA, but revise specific provisions for which the breach in the critical assumptions is relevant. For example, an APA may cover a series of controlled transactions for which the agreed methodology for only one specific transaction required revision.

In other circumstances, more general revisions may be required, depending on the specific facts and circumstances of a case, where allowed by domestic law. For example, tax administrations could consider evaluating the results of the transfer pricing methodology specified by the APA over the period of the APA, rather than on an annual basis, i.e. a “term test”. This approach would aggregate the financial results of FY2020, which may be exceptional, with the more normal results of prior and future years. In combination with this, tax administrations could also consider extending the period covered by an APA. Tax administrations could consider segregating the terms of the APA between financial years affected and unaffected by COVID-19, or cancelling an existing APA for FY2020 and then renewing the APA in a future period, potentially on revised terms. Alternatively, tax administration could also consider aggregating the financial results from a series of covered transactions covered separately under an APA and assessing whether on aggregate they deliver an outcome consistent with the terms of the APA.

In the context of the COVID-19 pandemic, there are two situations which may lead to cancelling an APA when it is established that: in FY2020 (i) there is a material breach in an APA’s critical assumption as a result of a change in economic circumstances; or, (ii) the taxpayer failed to materially comply with any term or condition of the APA. Cancellation would not, however, be automatic and the tax administration may waive cancellation under certain circumstances.52 Cancellation would have the effect of ending an APA on an agreed date or from a particular tax year or accounting period (e.g. after FY2019). 52 Paragraph 81 Annex II to Chapter IV of the OECD TPG.

Revocation may be considered where: (i) there is a misrepresentation, mistake or omission that was attributable to the neglect, carelessness, or wilful default of a taxpayer when filing an APA request and submission, the annual reports, or other supporting documentation or in supplying any related information; or (ii) the participating taxpayer (or taxpayers) fails to materially comply with a fundamental term or condition of the APA. The pandemic has not altered that standard. Accordingly, revocations of APAs whose terms extend into the period of the COVID-19 pandemic should be limited (just like arising in any other circumstances) to situations where the actions, as provided above53, that meet the standard for revocation regardless of whether such actions arose because of the pandemic. 53 Paragraph 77 Annex II to Chapter IV of the OECD TPG.

The timing for addressing the terms of APAs impacted by COVID-19 is important. Where material changes in economic conditions lead to the breach of one or more of the critical assumptions, taxpayers should notify the relevant tax administrations as soon as practicable after the change occurs, or the taxpayer becomes aware of the change. Early notification is encouraged in order to give the affected parties more time to try to reach agreement on revising the APA, thereby reducing the likelihood of cancellation.

As economic conditions will remain uncertain until at least the end of 2020, where taxpayers notify tax administrations of the failure to meet critical assumptions, tax administration may want to consider waiting for a reasonable period until data and information on the magnitude and longevity of the economic impact of COVID-19 are available before determining how to respond to a breach. By deferring their response, until more data and information is available tax administrations may find it easier to revise, rather than cancel, an APA.

Where the critical assumptions of an APA are breached, it is important that taxpayers collect and provide tax administrations with relevant supporting documentation. Depending on the particular critical assumption and other facts and circumstances, this could include, but may not be limited to: A description of the narrowest relevant taxpayer business segment tracked by management that encompasses the entities and covered transactions involved in the APA. Forecast and actual business segment profits for the financial years ending with or within financial years affected by COVID-19. Copies showing any proposed or implemented modifications to pre-existing agreements or of new intercompany contracts among the controlled parties affecting the covered transactions. A narrative explaining the anticipated effects of the current economic conditions on an agreed transfer pricing methodology during the financial years affected by COVID-19 including whether it caused restructuring of its operations and/or changes in its risks and responsibilities, and any mitigation of the impact of the current economic conditions on the tested party by government actions or other mechanisms such as business interruption insurance. The explanation must adequately demonstrate that the impact is attributable to the economic condition during the COVID- 19 pandemic, as a mere change in business results may be caused by other factors in a taxpayer’s business and accordingly may not constitute a breach of a critical assumption. A detailed profit and loss statement (“P&L”) with a breakdown of cost of goods sold (“COGS”) and selling, general and administrative expenses (“SG&A”) and other non-interest expenses for financial years affected by COVID-19 that include the covered transactions subject to the APA. The detailed P&L may include exceptional operating costs arising from COVID-19 or income from government assistance programme, including explanation on the accounting treatment of such costs or income. Information about third party behaviour.

When engaging with tax administrations, it is important that taxpayers are transparent and disclose all relevant information in a timely manner. In so doing, they will help to maintain the non-adversarial spirit and environment that is vital to the success of APA negotiations.

When considering the consequences of non-compliance with terms and conditions of an existing APA, tax administrations should adopt a similar approach to situations where there is a failure to meet critical assumptions. For example, when determining their response, tax administrations should consider (i) the terms of the APA; (ii) any agreement between relevant tax administrations as to how to deal with non-compliance; and (iii) any applicable domestic law or procedural provisions.54 54 Paragraph 74 of Annex II to Chapter IV of the OECD TPG.

However, tax administrations are likely to respond differently to the failure to comply with the terms and conditions of an existing APA, than to the failure to meet critical assumptions.55 This may reflect differences in the procedures prescribed by an APA agreement, domestic law or procedural provision. For example, it may be the case that where there is a breach in the critical assumption the terms of the APA prescribe that it should be cancelled unless otherwise agreed, whereas, the consequence of non- compliance may be that a tax administration can choose to cancel, revoke, revise or enforce an APA. 55 Section E.3 of Annex II to Chapter IV of the OECD TPG.

In the current environment, taxpayers may be reluctant about continuing or initiating new APA applications. This is understandable given the significant level of economic uncertainty that many businesses face, uncertainty that for some taxpayers may mean it is not feasible to reach agreements on future APAs today. However, it is important to acknowledge the role of APAs in securing tax certainty for taxpayers and tax administrations, and in preventing future tax disputes.

Where taxpayers and tax administrations are negotiating APAs that are intended to cover FY2020, all parties are encouraged to adopt a flexible and collaborative approach to determine how to take into account the current economic conditions, and the various options discussed above in relation to the revision of existing APAs will be relevant. For example, consideration could be given to agreeing a short period APA covering the period affected by the COVID-19 pandemic and a separate APA covering the post-COVID period. Another solution could be to conclude the APA for the whole period (e.g. APA period of 2020-2024) with a condition that the relevant impacts of the COVID-19 pandemic will be analysed and reported annually once they are known, and retrospective amendments to the APA made accordingly, when appropriate. Another solution could be to extend the period of the APA to mitigate the short term effect of the pandemic, depending on the magnitude and the length of such effect. Additionally, the use of a cumulative or term test throughout the APA period could be given consideration. In this context, it is important that taxpayers be transparent and disclose all relevant information concerning the impact of the COVID-19 pandemic on the covered transactions in a timely manner. The supporting documentation as mentioned in paragraph 104 of this guidance may serve as reference in this regard.

The COVID-19 pandemic also presents practical difficulties for tax administrations and taxpayers currently negotiating APAs. Potential challenges may arise for a number of reasons including restrictions on domestic and international travel; enforced or voluntary working from home; or additional resource pressures as a result of efforts to manage responses to the COVID-19 pandemic.

Despite these potential challenges, the value of achieving advanced certainty and effective dispute prevention through APAs remains compelling. A number of tax administrations and taxpayers have identified a variety of ways to adapt working practices in order to overcome any practical impediments to working APAs. Tax administrations and tax payers should recognise that rigid adherence to pre COVID- 19 working practises may unduly lead to significant delays in APA negotiations. Instead, innovative and flexible approaches to ensure collaborative working in order to minimise delays in concluding APAs under negotiation should be encouraged.

A range of technological solutions are available to replace and/or complement, traditional methods of communication, such as face-to-face meetings and the exchange of physical documentation, while maintaining confidentiality and security requirements. There may be situations where flexible approaches are not readily identifiable or appropriate, however, tax administrations and taxpayers should be encouraged to assess alternative approaches to maintaining progress in APA discussions on a case-by- case basis rather than defaulting to historic working methods or one-size fits all solutions. Effective use of tax administration and taxpayer resources is crucial during the period of the COVID-19 pandemic and practical experience suggests that the following have been utilised with success in certain cases: Virtual tax administration and taxpayer case conferences in the place of physical meetings (for example, telephone and video conferences) Virtual functional interviews with taxpayer’s employees Virtual taxpayer site visits in the place of physical visits (so long as the performance of adequate due diligence is not compromised) Electronic documentation sharing (via encrypted emails or electronic sharing platforms).