In the U.S., transfer pricing benchmarking under the Comparable Profits Method (“CPM”) or Transactional Net Margin Method (“TNMM”) depends on the availability of public company financial data. In recent years, the decreasing number of U.S. listed and non-exchange traded companies has made this benchmarking more challenging, not only due to the smaller population from which the comparable can be selected: Many of the remaining listed and non-exchange traded companies are either large companies that own intangibles or small companies that often operate at a loss. This trend should prompt transfer pricing practitioners to consider new, creative approaches in selecting comparable companies for purposes of CPM/TNMM, and in appropriate cases, to re-consider transactional or other methods that do not rely on publicly available profitability data. Further, an APA might now be a prudent choice to obtain certainty, even if APAs had not been considered necessary or worthwhile from a cost-benefit perspective in the past to mitigate tax risk.

Academic literature has acknowledged and examined the root causes of the recent decline in U.S. listed companies. For example, in their May 2015 working paper entitled “The U.S. Listing Gap,” Craig Doidge, G. Andrew Karolyi, and René M. Stulz investigated firms’ decisions to list and delist on public exchanges. The authors reported that the total number of U.S. listed companies fell from 8,000 to 4,100 from 1996 to 2012, while the rest of the world saw an increase from 30,700 to 39,400. The authors called this the “U.S. listing gap” and investigated possible explanations for this trend. Companies’ decision as to whether or not to “go public” is influenced by many factors, including prevailing economic conditions, government regulations, firm characteristics, and listing requirements on the major stock exchanges. To explain the gap, the authors investigated both the fall in the new list rate and the rise in the delist rate. They showed that the low number of U.S. listings from a global perspective was not just due to few IPOs, but it was also due to many delists. The authors ruled out industry changes, changes in listing requirements, and the reforms of the early 2000s as explanations for the gap. Instead, they showed that the delist rate rose because of an increase in merger activity involving publicly listed targets. This trend has also been noted by the Financial Times, which observed that “private equity firms and acquisitive companies have gobbled up many public groups. At the same time, ample venture capital and buoyant debt markets have allowed other fast-growing groups to stay private for much longer than in the past.”

The latest count of the U.S. listed companies, based on The World Bank data for 2018, is 4,397. However, some observers point out that this number conceals a more complicated trend: while the number of exchange-listed companies has gone down, the number of non-exchange traded companies has increased. “The number of securities that trade on OTC Markets, the alternative trading system for securities that don’t trade on an exchange, reached roughly 11,500 in 2020, more than double the total from 10 years ago.” Of these 11,500 companies, approximately 7,500 companies are based primarily in the U.S. or Canada, which would have been a healthy addition to the population of companies available for transfer pricing analysis – except that a closer look reveals that only about a third of these companies had financial data in at least one year during the 2017-2019 period. Further, of these companies, about one half were very small (less than $10 million in annual revenue) and almost 60% reported operating losses on average between 2017-2019, thereby leaving just over 1,000 companies, in addition to the 4,400 listed companies, which can be used in transfer pricing analysis of a North American tested party.

For a transfer pricing practitioner, the shrinking number of U.S. listed and OTC companies is creating a problem. Unlike the population of European companies, which are available from Orbis or Amadeus databases and comprise both publicly traded and privately held companies, the U.S. comparables are almost exclusively drawn from listed and non-exchange traded companies. This is largely due to the fact that most U.S. private companies are not required to publicly disclose their financial results. Of the approximately 10,700 private companies incorporated in the U.S. or Canada and included in the CapitalIQ database, less than 500 companies reported operating profit/loss data for all three years during the 2017-2019 period (about 200 companies reported the data for 2 years, and about 200 more companies had the data for one year). The smaller the available population of independent companies from which to select the comparable set, the greater the chance that any selected company will have subpar comparability to the tested party.

But the dwindling population is only a part of the problem. The companies within the U.S listed and non-exchange traded population are increasingly polarized by size, profitability, and geographical segments. Smaller companies often do not pass the size screen; they also often operate at a loss and may be rejected for benchmarking purposes because protracted losses can be viewed as being inconsistent with profit-seeking behavior of market companies. Larger companies frequently own significant intangible assets and have at least some portion of their operations outside the U.S. market – which throws comparability into question, particularly when the non-U.S. market is not just the place where the product is sold, but is also the location of the manufacturing or research facilities.

In other words, the size and the composition of the U.S. population of listed and non-exchange traded companies makes the circumstances in which U.S. “comparables” operate less “similar” to those of low-risk U.S. tested parties. This is a troubling development for the application of the CPM/TNMM, which has been a trusted workhorse in transfer pricing analysis for the past several decades. Similar challenges may also arise with respect to the residual profit split method (“RPSM”), which often applies CPM/TNMM analyses to benchmark routine contributions as an initial step prior to splitting non-routine residual profit.

Regardless of which judgement call is made, there will be uncertainty as to whether it will be questioned by both the Internal Revenue Service (“IRS”) and the tax authority of the foreign counterparty to the transaction. This makes robust transfer pricing documentation (in accordance with Internal Revenue Code section 6662 in the U.S., and local file requirements in most other jurisdictions that have adopted the OECD’s tiered approach) all the more essential as a first line of defense. Further, taxpayers faced with making these difficult judgment calls may wish to consider proactively requesting a bilateral APA, even if an APA was considered unnecessary or not worthwhile from a cost-benefit standpoint in the past when the CPM/TNMM could be applied more reliably due to a greater abundance of publicly available financial data for U.S. companies. In the current environment, taxpayers may now find the bilateral APA process to be an invaluable tool to mitigate the double tax risk in the U.S. and the foreign counterparty’s jurisdiction associated with any changes to the taxpayer’s application of the CPM/TNMM or adoption of a different method that may be implemented to address the challenges associated with diminishing U.S. comparables sets.