Harlow Higinbotham is a managing director and Vladimir Starkov is a director with NERA Economic Consulting in Chicago, and Nihan Mert-Beydilli is an associate director with NERA in Los Angeles.
The views expressed in this article are solely those of the authors and do not necessarily represent those of NERA Economic Consulting.
In this article, the authors examine how businesses can use their intercompany transfer pricing systems to better manage losses resulting from the coronavirus pandemic.
Business leaders and tax practitioners are considering how to address the serious challenges that may arise from disruptions in commercial activities caused by business closures, travel restrictions, and stay-at-home orders put in place in many countries to reduce the spread of COVID-19. Although this is not the first economic downturn many practitioners have experienced, it has numerous unique features, including supply- and demand-side elements, that have different effects on various economic sectors. Many industries stand to experience extraordinary business volume effects during 2020 caused by the prolonged disruption to daily life, interruptions in supply chains, job losses, and potential erosion of consumer confidence. Longer-term recessionary effects in 2021 and beyond, if any, are unclear. Among the questions companies will have to address is how to manage those losses using their intercompany transfer pricing systems.
Is an Adjustment Warranted?
Although this may sound like a truism, taxpayers should avoid making decisions regarding their transfer pricing arrangements that may not be seen as prudent after the crisis is over, including when those decisions are examined on audit years later. Any changes should be made after careful consideration and analysis of the facts, circumstances, and available data.
How should taxpayers decide whether transfer pricing adjustments are warranted in the first place? In transactions between affiliates that the company’s transfer pricing policy treats as routine on one side and entrepreneurial on the other, the decision whether to adjust the results of the policy depends on the actual roles that routine and nonroutine affiliates performed during the downturn. For example, facts about how proactive the routine affiliate was in mitigating the impact of the economic downturn should be examined. It is crucial to document those facts as close to the timing of the events as possible and incorporate them either into transfer pricing documentation or into an audit defense file. In transactions in which joint profit is shared among at least two affiliates (for example, under a profit-split method), the question to be examined is how independent parties would split the losses caused by an event not under their control in the context of the joint venture implicit in the profit split. For example, one query would be to determine whether arm’s-length parties would make retroactive profit adjustments or would instead have agreed on prospective adjustments.
Assuming the decision to proceed with adjusting transfer prices has been made, the next step is to consider what tools to use to accomplish it. This article describes only some of the methods that may be used. Selection of any particular method and its application should be based on analysis of the taxpayer’s facts and circumstances together with an assessment of the available evidence from relevant third-party benchmarks and comparables.
Intercompany Pricing
While the taxpayer’s financial results and forecasts will reflect the current economic conditions, the same is not necessarily true for the available benchmark comparable company data. That may be because some potential comparables are not affected in the same way as the taxpayer, or because up-to-date results for comparables are not yet available. Continued use of the benchmark comparables’ data that are not synchronized with the taxpayer’s time period presents serious challenges to taxpayers in terms of setting prices for the current year, budgeting for coming years, and complying with applicable transfer pricing regulations. To ensure that potential benchmark comparable company data are indeed comparable and reflect the actual economic circumstances for the related parties, several adjustments can be considered.
Comparisons Over Multiyear Periods
The one-sided transfer pricing methods presuppose that returns for routine functions are to be determined by reference to a normal range of returns for comparable entities over one year or several years. The normal range of returns is typically understood as the interval comprising the second and third quartiles of the range of returns calculated from the benchmark comparables’ financial results achieved during normal economic times. However, when entire sectors of the economy are unable to do business as usual, the interquartile range of comparables’ results achieved in prior periods will unlikely be a sensible benchmark. If multiple years of data are used to calculate an arm’s-length range (as would be the case under U.S. regulations), period-average results for the selected profit level indicators need to reflect the estimated benchmark results of those indicators including 2020, and the benchmark results observed exclusively for rise in importance in 2020 as indicators of arm’s-length returns in that year.
Using Forecasted Financials for Comparables
When the economic conditions in which the comparables achieved their profit levels are materially different from the economic conditions facing the tested party, it may be appropriate to consider using current-year forecasted results of the comparables to calculate the arm’s-length range. Moreover, companies that are good comparables in normal economic circumstances may not be closely comparable in the COVID-19 downturn because of the different economic effects noted above. Also, there are implementation challenges stemming from the availability and reliability of up-to-date financial forecasts for the available benchmark comparable companies. Even so, this approach may still provide a useful indication when setting prices for budget purposes.
Reexamining the Set of Comparable Companies
In a sharp and unexpected revenue decline, the key factor that affects a company’s profitability (for example, the magnitude of the operating loss) is the proportion of fixed costs to variable costs. Unlike variable costs that fall in proportion to a decline in revenue, fixed costs cannot be reduced quickly. Therefore, one of the key comparability factors in an economic downturn will become the ratio of the comparables’ variable to fixed costs. The comparable companies selected for calculating the prior arm’s-length range can be reexamined to isolate those that have experienced similar levels of sales decline or that most closely match the tested party’s ratio of variable to fixed costs, and the arm’s-length range can be recalculated using only that subset. Alternatively, when reliable data for comparable companies are not available, a pro forma analysis of the related-party financial results may provide suitable evidence of the anticipated or budgeted results — that is, “but for” the downturn — to demonstrate and document their arm’s-length character.
Adjusting the Historical Comparables Data
A regression analysis can be used to determine the systematic differences in a given profit level indicator in past years of observations, and an adjustment for the current period can be applied to historical financial results of the comparable companies using this systematic component. Other regression-based analyses can be performed to measure the systematic relationship between changes in sales and profitability of the comparables or to account for the differences in the ratio of fixed costs to total costs among the comparables. That can be done using existing comparable sets, depending on sample size and reliability of the resulting coefficients, or expanded sets. The resulting regression coefficients provide an estimate of the decrease in profitability (as a percentage of sales or assets, for example) for each percentage of decrease in sales volumes and can be used to adjust each comparable company’s profitability based on the percentage decline in the taxpayer’s sales. Historical financial data of the benchmark comparable companies recalibrated using regression analyses can then be used to derive the adjusted arm’s-length range.
Managing Systems Based on Profit-Split Methods
The conceptual paradigm underlying the profit-split method is a joint venture between two or more parties, each of which makes both routine and nonroutine contributions to the venture.
The paradigm implicit in the profit-split method is the assumption that the joint venture partners contemplate a normal range of business risks and potential business outcomes. The profit-split method, at least as delineated in U.S. Treasury regulations and the OECD transfer pricing guidelines, does not address how the joint venture partners would or might respond to unforeseen events that dramatically change the economic environment of their business.
Except possibly for force majeure clauses, which excuse performance in the event of such natural or manmade disasters as hurricanes and wars, generally there is no typical contractual provision in arm’s-length joint venture agreements governing the consequences of unforeseeable events that materially affect the economic circumstances of the venture. Even so, for transfer pricing purposes it is necessary to postulate how the joint venture partners would decide, either ex ante in the joint venture agreement or ex post in amendments to the agreement, to modify the terms of their venture to take into account unforeseen events. While there is no single course that joint venture partners in general might be predicted to choose, there is one that in the context of profit splits, parties at arm’s length could reasonably adopt ex ante or ex post.
The question that arises in extraordinary economic conditions is how extraordinary risk should be allocated among routine and nonroutine functions. Because the implicit agreement imposed by the residual profit-split method under normal circumstances is for the routine functions to bear risk in the interquartile range of comparables’ rates of return and for the nonroutine functions to bear the residual risk, a logical extension of that agreement in extraordinary conditions would expand the risk borne by the routine functions beyond the normal, interquartile, range of returns for comparable routine functions but not beyond the full range of returns observed for comparable routine functions, which includes observed extreme returns as well as normal returns. While the central 50 percent of the comparables’ range (the median) may be appropriate in ordinary circumstances, part or all of the other 50 percent of the comparables’ range should be feasible in extreme circumstances. Further, the adjustments described above for applying one-sided methods all apply for determining the returns for routine contributions under the residual profit-split method.
Alternatively, losses associated with unforeseen events can be isolated and split on a different basis than normal operating results would be. In a residual profit-split context, without any adjustments to the routine benchmarks, those losses would be attributed to the nonroutine, entrepreneurial activities and would be split between the entrepreneurs. The extraordinary losses resulting from a substantial decline in production and sales arguably result from risks borne by the entire business, not just the entrepreneurial activities. They should therefore be allocated separately based on a measure that represents both routine and nonroutine activities. For example, in a profit-split system in which residual profits are shared based on capitalized costs of entrepreneurial activities that is, entrepreneurial assets — the extraordinary losses can be allocated by total company assets, including both the routine and nonroutine, entrepreneurial assets.
The important step in implementing that allocation is to quantify the losses attributable to extraordinary events, which would include all direct costs readily attributable to addressing the unforeseen events and indirect costs or lost profits caused by disruptions in economic activity. One way to measure the impact on profitability is to examine the difference between budgets prepared before and after the unforeseen events or between budgets prepared before the events and the actual year-end financial data for the business.
What to Do Next?
Although the methods to address transfer pricing challenges brought on by this economic downturn will be specific to each taxpayer, there are several practical action steps all taxpayers should take:
determine whether a change is required to current transfer pricing policies;
perform an analysis to determine losses associated with the extraordinary events and determine which entities will bear associated risks;
in anticipation of the preparation of transfer pricing documentation, discussions with tax authorities, and potential disputes, carefully document contemporaneous evidence, considerations for any changes in transfer pricing policies, and any analyses undertaken;
reexamine the suitability of each of the comparable companies used in the existing transfer pricing system and prepare adjustments that may be appropriate to account for the effect of extraordinary events; and
when transactions are covered by advanced pricing agreements, be prepared to fully disclose to the relevant tax authorities the facts and circumstances that affect operations and financial results of the taxpayers, and present proposals for any amendments to the agreements.