Testing the meaning of the arm’s-length principle in 2020 and beyond

The COVID-19 pandemic is the defining event of 2020. Countless articles have been and will be written documenting the continuing impact of the coronavirus on the global society and economy. As governments gradually reopen their economies, many countries are concerned that a second wave of infections may slow the economic recovery until 2021 or beyond.

The OECD Economic Outlook, released in June 2020, indicates that regardless of whether there is a 'single wave' or a 'double hit', the decline in output and the rise in unemployment will be significant. Global GDP is expected to fall by 6% in 2020 and unemployment is expected to rise to more than 9% in the single wave scenario, and even in that case, the global economy is not expected to return to pre-crisis output levels by the end of 2021.

Over the next several years, national governments will likely need to collect additional tax revenues from various sources to pay for the programmes and investments put in place to support the recovery. In light of all this, transfer pricing (TP), which multinational businesses use to calculate their income tax liabilities in the jurisdictions in which they operate, and which is enforced globally based on the application of the arm's-length principle, may be impacted as governments seek to ensure that they collect the appropriate amounts of tax revenues.

Article 9(1) of the OECD Model Tax Convention on Income and on Capital defines the arm's-length principle as follows:

Many multinational businesses have structured their operations to comply with the arm's-length principle by organising networks of 'principal' companies and 'routine' companies. In such structures, the principal companies enter into inter-company arrangements requiring that they conduct the important entrepreneurial functions and bear the entrepreneurial risks in return for the entrepreneurial profits and losses of the global business. The routine companies enter into inter-company arrangements requiring them to perform manufacturing, distribution or service functions and do not bear the entrepreneurial risks in return for earning routine profits (or sometimes losses) determined by reference to comparable companies.

Over the past decade of strong to moderate economic growth, it has been our observation that such structures have resulted in principal companies earning the entrepreneurial profits and losses of the business, while the routine companies earned modest profits benchmarked to comparable companies.

The COVID-19 health crisis has caused much of the world to enter a state of social lockdown in an effort to slow the spread of the disease, and this has resulted in a significant decline in global GDP. It is likely that many multinational companies will report losses for 2020. Typically, companies that incur significant losses are expected to cut costs to improve their financial viability.

However, the speed with which COVID-19 spread globally and the suddenness of the resulting lockdowns prevented companies from adjusting their costs. Even though many businesses have temporarily closed or reduced their operations, they are incurring ongoing losses, such as:

• Distributors experiencing significant sales declines but not being able to reduce selling, general and administrative (SG&A) expenses proportionately;

• Retailers shutting down stores due to public health restrictions or reduced demand but still incurring rent, distribution network, and other costs;

• Manufacturers shutting down or slowing operations due to lack of demand, supply disruptions or social distancing requirements but still incurring depreciation expenses, carrying and maintenance costs; and

• Service providers shutting down or working at reduced capacity due to lack of demand but still incurring facilities and staff costs.

Given the above factors, it seems likely that many companies will incur losses in 2020 if their fixed costs exceed their income due to the significant reduction of business volume. Tax authorities seeking revenues to pay for programmes and investments necessary to support the COVID-19 recovery may start TP examinations for 2020 relatively soon and thereby test the application of the arm's-length principle during a period of extreme economic stress. The statute of limitations on tax assessments in most countries is several years and often longer, suggesting that COVID-19 will likely impact TP examinations for many years after the pandemic has abated.

On June 3 2020, OECD's Centre for Tax Policy and Administration, Committee on Fiscal Affairs, Working Party No. 6 issued a questionnaire to businesses (see CTPA/CFA/WP6/ NOE2(2020)6)) requesting comments on the application of the arm's-length principle in the COVID-19 impacted global economy. The introduction to that questionnaire (page 2) anticipates some of the challenges going forward:

In this regard, the OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD TPG) continue to represent internationally agreed principles and, irrespective of the underlying economic circumstances, provide guidance for the application of the arm's length principle of which Article 9 is the authoritative statement. However, members also recognise that in these challenging times it is incumbent on us to consider what additional tools could be provided for greater tax certainty and to mitigate the risks of future disputes between taxpayers and tax administrations."

Globally, TP examinations conducted over the next few years will likely test the meaning of the arm's-length principle by raising the following issues:

During the past decade of economic growth, tax authorities have become accustomed to routine entities earning a modest profit based on comparable company benchmarks. They may not be willing to accept that the inter-company agreements do not guarantee a modest profit in all circumstances, and that the COVID-19 pandemic is an extraordinary event for which parties transacting at arm's-length would agree to share losses.

Multinational companies will be looking to prove that uncontrolled companies used for benchmarking report lower results and even losses in the crisis environment (although the inclusion of such companies as comparables typically has been controversial). The lower than usual results reported by tested entities during the crisis might call for specific considerations on how to allocate those losses within the multinational enterprise.

The OECD TPG do not address directly arm's-length behaviour in times of crisis. They do provide, however, a well-established approach to the application of the arm's-length principle and indicate that loss-making comparables that satisfy the comparability analysis should not be rejected on the sole basis that they suffer losses.

In times of prosperity, based on our observations, acceptance of loss-makers as comparables has been problematic. In times of crisis, if companies satisfying the comparability criteria report losses or lower than typical results, it seems reasonable that adjustments should be allowed to reflect those lower results or losses in the benchmarking analysis relevant for the crisis period(s). The important factor for acceptance or rejection of companies as comparables should be whether they are comparable (i.e., meet the comparability criteria) and not whether they are loss-makers.

Taxpayers preparing TP documentation should provide qualitative and quantitative information to explain the reasons for their COVID-19 related losses. Taxpayers should begin assembling this information, because it may be difficult to do so in the future, once a TP examination has started. Many countries do not allow post-year-end adjustments but the comparables data to support adjustments will not be available until many months after the year-end.

Conducting a benchmarking analysis for the years impacted by COVID-19 may be challenging as the financial data typically used for benchmarking studies only becomes available after a lag period of one to two years. This means that at present, financial data for 2018 may be the most recent available point of reference for the taxpayer. As the use of data for years of economic prosperity to conduct analysis for crisis years is intuitively inappropriate, taxpayers may find it difficult when planning and documenting FY2020.

A thorough TP analysis, including reviewing the inter-company agreements, the function and risk profiles, and conduct of the parties during the COVID-19 period may be required by tax administrators to verify any changes in the TP policy and/or the arm's-length nature of losses. At the same time, it seems reasonable that the authorities should not propose a TP assessment merely because a controlled party classified as routine reported a loss.

In terms of availability of comparable data, the tax authorities might have the benefit of hindsight: as TP audits will be conducted several years after the end of 2020 (and, perhaps, other COVID-19 impacted years), the financial data for those years would be available for conducting the audit-related analyses.

Two important questions will need to be asked in this context:

• Is it appropriate to use – for audit assessment purposes – data that was not available to the taxpayers when they adopted and documented the terms of inter-company agreements?

• Is it appropriate to use data that reflects the situation of only those entities that survived the pandemic? What about entities that either go bankrupt or are taken over, for which data will no longer be available in public databases?

Experience from the 2008-2009 financial crisis showed that many companies with weak financial results went bankrupt or were acquired by competitors. As a result, some profitability ranges assembled in the wake of the crisis did not reflect the expected decline in financial results.

Accordingly, adjustments to benchmarking ranges might be required in order to benchmark the crisis. Benchmarking approaches may include:

• Expanding comparables to include loss-makers that satisfy comparability criteria in order to better capture the COVID-19 impact;

• Narrowing comparables to exclude companies with significantly different cost structures (operating leverage/ fixed cost ratio) to improve reliability of data; and

• Adjusting comparable data to reflect the impact of COVID-19 on profitability in 2020 and creating specific COVID-19 ranges.

In the optimistic hope that the COVID-19 pandemic will abate and global society and the economy will recover and thrive, if companies report improved profitability in future years before a TP examination starts, should this post-year end information be taken into account in examining an earlier year?

Paragraph 3.75 of the OECD TPG notes that examining multiple year data is often useful in a comparability analysis, but it is not a systematic requirement. The guidelines go on to note that multiple year data should be used where they add value to the TP analysis.

Tax penalties are typically enacted to encourage compliance with the TP rules. The OECD TPG make two points regarding penalties that are particularly relevant to the COVID-19 pandemic.

First, the imposition of a sizable "no-fault" penalty based on the mere existence of an understatement of a certain amount would be unduly harsh if the understatement is attributable to good faith error rather than negligence or an actual intent to avoid tax.

Second, it would be unfair to impose sizable penalties on taxpayers that made a reasonable effort in good faith to set the terms of their transactions with associated enterprises in a manner consistent with the arm's-length principle. (See OECD TPG, paragraph 4.28.)

These notes suggest that tax authorities should exercise restraint in applying penalties when losses are primarily driven by the COVID-19 pandemic. In addition, as discussed below, taxpayers often request Mutual Agreement Procedure (MAP) following a TP assessment to eliminate double taxation. Penalties are generally not subject to MAP relief, however, and the imposition of penalties may unnecessarily complicate negotiations to eliminate double taxation.

TP audits and adjustments made for COVID-19 impacted years could result in higher double economic taxation worldwide. An expected response to that would be the increased use of advance pricing agreements (APAs) and MAPs. Competent authority assistance for double taxation is provided under the MAP article of the relevant tax treaty, and in the case of European Union member states, also under the EU Arbitration Convention (AC) or the EU Dispute Resolution Directive. To provide relief from double taxation, the respective competent authorities must be notified of the proposed TP adjustments, or a request for MAP assistance must be filed, within specified deadlines.

Increased use of MAPs, as well as APAs, are expected as a result of the pandemic, as companies seek to resolve and avoid TP disputes.

During the 2007-2009 financial crisis, the competent authorities considered 'down economy adjustments' on a case-by-case basis. Whether such adjustments were made depended on a variety of factors, including:

• Whether the tested party and the comparables experienced similar effects from the downturn;

• The tested party's historic risk profile and performance; and

• The taxpayer's willingness to accept a symmetrical adjustment (e.g. in a renewal APA) when the economy improves.

Approaches considered included:

• Changing the APA term;

• Waiting for more up-to-date financial data; and

• Using a different set of comparables, and/or applying a longer testing period.

In an environment reflecting the impact of COVID-19, taxpayers with active or pending APAs may be advised to revisit the critical APA assumptions and to align past years, the COVID-19 period, and future years. Disclosure of relevant facts concerning transactions already covered by an APA and proposing appropriate amendments to the APA would be important to maintain the validity of the APA in 2020 and post-COVID-19 years.

2020 is a unique year. Although the pandemic will likely impact the business models of multinationals in all industries in different ways, multinationals across industries need to assess the disruption to their businesses and its impact on their TP policies. Moreover, companies need to consider whether benchmarking adaptations and adjustment of routine returns and profit/loss split allocations are required.

The impact of COVID-19 will likely be much broader than 'just financial' and in many businesses that impact will be felt throughout the entire value chain both during the COVID-19 crisis and under the new normal.

Click here to read the entire 2020 Deloitte/ITR Transfer Pricing Controversy guide