 |
| Alan Alford |
Between the third quarter of 2008 and the second quarter of 2009, the U.S. economy contracted by 3.25%.[1] Other countries had similar experiences with a sudden contraction in their economies because of the financial market illiquidity.
Second, the worldwide spread of COVID-19 in 2020 has produced an even stronger shock to the economy as workers worldwide have self-quarantined to reduce the disease's spread. Initial estimates predict a 4.8% contraction of U.S. gross domestic product in the first quarter of 2020.[2] As of May, estimates for the second quarter forecast an even deeper contraction.[3]
The current economic crisis will have a severe impact on all companies. Almost all are preparing for seriously reduced profits, if not substantial losses, because of stressed supply chains and shifting demand for their products and services. Many have withdrawn their public guidance because of the uncertain business environment.[4]
Chief financial officers and their tax departments are reviewing their companies' internal forecasts to understand how this contraction will influence their tax liabilities. A typical multinational company structure includes at least one principal entity that bears companies' overall business risks and own their key intangibles and valuable assets; these entities will absorb the majority of the multinationals' decline in profitability.
Multinationals also include several low-risk entities in their operational structures that are used to test the arm's-length nature of the companies' transfer pricing policies. These service providers typically earn a steady, predictable return, independent of the principal's commercial risks. They are, however, not isolated from the broader economic risks.
These entities will see a reduction in profitability because the factors that drive their profits — demand for their services — will decline significantly. Tax directors should review their comparables — a set of third-party companies or transactions that are used to benchmark a company's particular related-party transaction — to adapt their transfer pricing policies to the COVID-19 economic shock and the coming years as the world economy recovers.
In this article, I look broadly at three groups of companies commonly used to benchmark related-party transactions — distributors, manufacturers and service companies. First, I examine these groups' profitability subsequent to the Great Recession. Second, I consider how these lessons may be applied to the current environment for related-party transactions.
Based on experience from the Great Recession and its aftermath, these routine service providers will also experience a decline in their profit levels over the coming years. Many of these companies could report operating losses, even though they bear low risks.
The severity and expected length of the COVID-19 shock will likely produce a greater impact on the arm's-length pricing for these transactions than the Great Recession did. This will cause an even stronger reaction to the comparables that companies use to benchmark their various related-party transactions. As a result, the results discussed in this paper will likely be even stronger and more sustained during the aftermath of COVID-19 than they were after the Great Recession.
Developing Broad Sets for Routine Returns
I develop three sets of companies using the criteria shown in Table 1. These searches would be the foundations for any comparable search to benchmark the routine returns. Each set starts with a selection of standard industrial classification codes typically used to select comparables for that type of related-party transactions, located in the U.S. or Canada.
Next, I focus on those companies for which data are publicly available in S&P's Capital IQ — either public companies or companies that have publicly traded debt — to ensure a reliable source of financial information.
I eliminate companies whose auditors reported going concern issues between Jan. 1, 2003, and Dec. 31, 2007 — the precrisis period — as their results would likely be driven by issues other than those in the general economy.[5]
I also eliminate those companies that have research and development-to-sales or advertising-to-sales in excess of 3% during the precrisis period; expenditures in excess of these levels tend to indicate the creation of valuable intangibles that these routine service providers typically do not have.
Finally, I eliminate companies that have extreme returns, measured by operating margins less than -15% or greater than 20%. These companies likely reflect startups or other issues that skew the reported margins.[6]
Table 1: Summary of Search Criteria
Table 2 shows the interquartile range for each set in the precrisis period. A transfer pricing study would consider the interquartile range — the returns for the middle 50% of observations — to determine the arm's-length range that the related-party service providers should earn.[8]
To define the precrisis period, I focus on the companies' five fiscal years ending prior to Oct. 1, 2008, to ensure that the estimated returns reflect the run-up to the 2008 crisis.
Table 2: Precrisis Period Operating Margins for the Sets (2003–2007)
Post-Crisis Performance for the Broad Sets
Firms with low risks can experience operating losses due to the general economic conditions. Table 3 shows that companies experienced more losses in 2009 than in 2008 as the full effects of the crisis affected the economy at large.
In each set, the number of loss companies at least doubles in 2009. For the five-year period, though, fewer companies exhibit losses, suggesting a potential transitory nature of the shock. Comparables will be directly affected by the shock but will eventually recover. Intermediate returns for the tested party necessarily have to reflect that reduced profitability, including the possibility of operating losses in a given year.
Table 3 introduces the possibility that a company may properly assign losses to its routine services during the COVID-19 crisis, and possibly for the succeeding years, because of the losses that the identified comparables will likely experience. It also indicates that any adjustment to the transfer pricing will likely not be consistent across the routine, low-risk services.
Table 3: Companies Reporting Operation Losses in a Particular Period
Table 4 presents the period-average returns, which are used to "reduce the effect of short-term variations that may be unrelated to transfer pricing" such as the economic shocks of the Great Recession.[9] As with Table 3, there are differences across the three sets. However, no set shows a decrease that might be expected from looking at the results shown in Table 3.
Table 4: Profitability Post-Crisis
Distributor margins do decline in the post-crisis period, but only marginally. In both post-crisis periods, the lower quartile is roughly the same. Unlike the other two set, the median and upper quartile for the distributors fall more than the lower quartile. The five-year period does show a higher return for both than the three-year period.
This would be consistent with companies returning to their precrisis profitability after the initial shock and suggests that the impact of the Great Recession was reversed after three years and profits returned to precrisis levels.
Manufacturers reported a substantial widening of the interquartile range. The upper end of the range is roughly the same, but the median and the lower end of the range fell by 1.7%. There is some recovery in the final two years, since the lower end of the quartile is higher, but it still remains 1% lower than the precrisis returns. It suggests that manufacturing companies did not recover quickly from the economic shock as quickly as distribution companies.
A transfer pricing analysis would reasonably reflect a slower recovery from the shock after COVID-19 than would distribution companies; depressed returns will likely continue for a longer period for manufacturing companies than for distribution companies.
Services, however, increase in the post-crisis period. Their three-year profitability is higher than their five-year profitability. Third-party service recipients did reduce their employment, likely increasing their demand for services. This is especially true at the higher end of the range, which continues to demonstrate a higher level of profitability over the final two years than in the precrisis period.
Individual Performance Pre- and Post-Crisis
The results in Table 4 could be influenced by the exit of companies that had poor performance precrisis; this is commonly known as survivorship bias. Those companies would likely have been weaker financially and less likely to survive the Great Recession.[11]
In this section, I concentrate on those companies that survived the Great Recession and reported returns in both periods. This analysis will give tax directors a more reliable indication of how their routine return policies will change over the coming years.
To enhance the comparability of the sets, I retain only those companies that reported financial returns for the precrisis period as well as the post-crisis period. This minimizes the chances of survivorship bias. It also provides insight into the question of how a multinational company's comparable sets would be expected to change in the post-crisis period.
This reduces the set further, as shown in Table 1. Table 5 shows how companies' profitability changed between the pre- and post-crisis period. As would be expected, the manufacturing and distribution sets are skewed toward lower returns post-crisis.
Table 5: Matching the Sets Across Period
"Exited companies" in the last row of Table 5 refers to companies that reported sufficient data in the pre-crisis period but not in the post-crisis period. These companies comprised more than a quarter of the surviving companies.[12] This would explain why the returns increase in Table 4 from Table 2.
This is an example of the survivorship bias, which would push up the returns and inappropriately slant the interpretation. Overall, other things being equal, the survivorship bias tends to overstate the returns that comparable sets would exhibit at arm's length. The transfer pricing analysis adapted to the terms and circumstances of the related-party transactions will reasonably show a more consistently lower interquartile range.
Another Effect on the Total Profits Attributed to the Routine Returns
The comparables' total profitability is affected by more than just their respective operating profits. Table 6 reflects that many of the companies incurred restructuring costs as a result of the economic environment.[13] Consistent with the results above, manufacturing companies experienced the most restructurings.
The total number of companies incurring restructuring charges does not equal the sum of charges in the individual years as some companies record these charges over multiple years. That practice would reflect continued issues post-crisis that necessitated multiple changes to how the third parties serviced their customers.
Table 6: Number of Companies Incurring Restructuring Charges Post-Crisis
These restructuring costs are not part of the companies' operating incomes. Instead, they are recorded as extraordinary costs that are independent of operations. Such charges include various impairments, closing of operations and severance costs that arise because a company realizes that it does not have sufficient demand for its services to justify those investments. As a result, the company sustains reductions in assets and other one-time charges. A company is not able to pass these charges to its customers, as they do not directly benefit the customers and are unrelated to the company's continuing operations. Total profits would be reduced because of these expenses.
Restructuring charges may arise in the related-party context as well. Much as with the comparables, these costs should be retained by the particular service provider and not shared with other related parties. This would further reduce the profitability of the tested parties beyond what had arisen from the returns in Table 4.
Implications for Transfer Pricing Policies
The COVID-19 shutdown has pushed the world economy into its deepest contraction since the Great Depression of the 1930s. Corporate profits will decline during 2020 and 2021. Tax directors will need to understand how this overall decline in profits will influence the companies' individual tax liabilities, including the profitability of certain routine services — the companies' distribution networks, their contract manufacturers and various other service providers within the corporate structure.
A basic tenet of transfer pricing is that related-party transactions should reflect the pricing of third-party transactions. As those third-party transactions are exposed to the general economic conditions, related-party transactions will be as well. By extension, if third-party comparables report lower profitability because of the general economic environment, the related-party service providers will likely also report a lower profitability.
The current economic environment parallels what happened during the Great Recession of 2008 and 2009. Then, corporate profits declined, including those that enabled the offering of services comparable to those provided by related parties. An examination of companies with the same general characteristics as those used in transfer pricing studies shows that those companies did generate less profit following the economic shock.
I draw three conclusions from this analysis.
1. Some comparable companies will reasonably report operating losses because of the COVID-19 crisis.
Economic shocks introduce the possibility that a related-party transaction will generate an operating loss yet be consistent with the arm's-length standard. Multiple companies reported actual losses from third-party transactions that are broadly similar to the related-party transactions structured by multinationals after the Great Recession.
In 2020 and 2021, the arm's-length range will indicate that the related parties could reasonably have operating losses just because of the magnitude of the contraction. Thus, a particular tested party could have an operating loss and still be consistent with the arm's-length standard.
The duration of any losses depends on the length of the shock. Losses from the Great Recession continued through 2009; five-year returns show increasing returns in the final years. That reflects the reasonably quick rebound post-recession. Economic growth restarted in the third quarter of 2009. By 2010, profitability began returning to precrisis levels.
The shock from COVID-19 is significantly stronger and may last longer. As a result, more comparable companies may report losses and may sustain those losses for longer periods.
2. Interquartile ranges are likely to show larger and more sustained drops than they did post-Great Recession.
Interquartile ranges fell post-Great Recession but appear to have recovered largely after 2010. The severity of the COVID-19 shock, and any extended duration of this crisis, will more deeply depress the returns for the comparable companies and extend the time it takes for them to recover to precrisis levels of profitability.
Comparable, related-party transactions will need to reflect the influence of the economic environment in which they operate, just as third-party companies do. Multinationals' transfer pricing policies will necessarily allocate a reduced level of profit to their low-risk entities for extended periods.
It will be incumbent on them to review their comparables' profitability on an annual basis to determine whether the revised policy should revert to precrisis policies.
3. Tested parties will bear extraordinary expenses that are excluded from a company's transfer pricing.
Additionally, nonoperational expenses will further reduce the related-party service providers' taxable incomes. Subsequent to the Great Recession, many companies incurred restructuring charges as they right-sized their businesses. At arm's length, these restructuring charges could not be charged to their third-party customers.
Similarly, in the related party context, the relevant service providers need to bear those charges directly. For the benchmark companies, these costs arise because the companies needed to change to serve their customers more efficiently. It is reasonable to believe that related parties, operating at arm's length, would not be able to recover the full value of these costs. Therefore, they should be excluded from any transfer pricing adjustments.
Conclusion
The COVID-19 economic crisis will negatively influence a multinational organization's profitability. That negative influence will be felt by the multinational's entities that are low-risk service providers within the corporate transaction structure.
The third parties used to benchmark the transactions will likely report lower profit levels for several years before reverting to precrisis levels. It is conceivable that many of these companies will report operating losses and indicate that related-party transactions could have operating losses as well.
Finally, the multinational may undertake certain restructuring transactions and incur other one-time costs to adapt to the new economic environment. Those costs reasonably remain with the entity that directly incurs those costs, just as they stay with comparable firms that incur similar charges.
Alan Alford, Ph.D, is a principal at Bates White LLC.
The author would like to thank Marina Gentile and Heather Dittbrenner for their helpful comments.
The opinions expressed are those of the author(s) and do not necessarily reflect the views of the organization, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.
[1] "Industry Data: Interactive Access to Industry Economic Accounts Data—GDP by Industry," accessed May 13, 2020, https://apps.bea.gov/iTable/iTable.cfm?ReqID=51&step=1.
[2] US Bureau of Economic Analysis, "Gross Domestic Product, 1st Quarter 2020 (Advance Estimate)," April 29, 2020, https://www.bea.gov/news/2020/gross-domestic-product-1st-quarter-2020-advance-estimate#menu3.
[3] St. Louis Fed, "Economic News Index: Real GDP Nowcast, Percent Change at Annual Rate, Quarterly, Seasonally Adjusted Annual Rate," https://fred.stlouisfed.org/series/GDPNOW.
[4] Publicly traded companies use this guidance to provide shareholders and other interested parties with expectations for future earnings. For example, Apple withdrew its 2020 guidance because of the issues with its Chinese factory. "Given the lack of visibility and certainty in the near term, we will not be issuing guidance for the coming quarter." (Apple Inc., Earnings call for FQ2 2020, April 30, 2020, p. 6).
[5] I purposely do not examine whether any of these companies had subsequent going concern issues as that could result from the general economic contraction.
[6] This screen would not typically be part of a benchmarking exercise. However, that exercise would include a rigorous qualitative review of the companies passing the quantitative review. Many of the companies with extreme returns are startup companies, have limited operations, or possess other characteristics that would distinguish them from most tested parties and exclude them from most transfer pricing analyses.
[7] A standard transfer pricing analysis typically uses the return on operating assets to benchmark the returns from a low-risk manufacturing set.
[8] See Treas. Reg. §1.482-1(e)(iii)(C).
[9] Treas. Reg. §1.482-1(f)(2)(iii)(D).
[10] If a company does not have three years of data post-2008, I consider it to have "exited' from independent operations and do not include it in Table 4 or subsequent tables.
[11] Similarly, COVID-19 will likely force many companies used as comparables in transfer pricing analyses to declare bankruptcy, even though they had no going concern issues prior to this crisis.
[12] Exited manufacturing and distribution companies are approximately 16% and 15% of the surviving companies, respectively.
[13] I note that these costs are extraordinary expenses, not an operating expense. Thus, these expenses would not be reflected in a standard transfer pricing calculation. A tested party may still report losses for tax purposes, even if the transfer pricing would give it a profit.
For a reprint of this article, please contact reprints@law360.com.