Arm's Length Principle vs. Formulary Apportionment in BEPS Action 13: Stakeholders’ Perspectives

ABSTRACT

This paper analyses comment-letter lobbying by different stakeholders to influence OECD’s documentation rules on transfer pricing within the Base Erosion and Profit Shifting (BEPS) project. Using a content analysis of stakeholders’ comments to a discussion draft on documentation requirements of Action 13 of OECD’s BEPS policies, we offer some evidence that professional accounting firms may have viewed the new documentation requirements as a stealth paradigm shift away from the arm’s length principle (ALP) to formulary allocation (FA). Our analysis suggests that other stakeholders from academia and civil society were sceptical of the comments expressed by the professional tax firms, regarding them as a means to resist changes in transfer pricing documentation. We suggest that professional accounting firms’ comments may have represented implicit lobbying against changes to the ALP regime that may have been considered beneficial to taxpayers in reducing their aggregate tax burden despite its implementation and rule complexity.

Keywords:

• transfer pricing
• arm’s length principle
• formulary apportionment
• Base Erosion and Profit Shifting (BEPS)
• Lobbying

Acknowledgements

Funding for this research from the Social Sciences & Humanities Research Council and the Public Interest Accounting Group at the Schulich School of Business is gratefully acknowledged. We also appreciate comments from anonymous reviewers of this journal.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1 FA also offers other benefits. For example, if the FA is based on payroll costs, then it could increase the labour component in the high-tax jurisdictions to maximize the tax shield for labour costs. It would allow intangible capital to be located at the parent company’s jurisdiction where the research personnel are likely to be located, instead of artificially locating high-value intangible patents in low-tax jurisdictions. If the FA is based on revenues, then the jurisdiction where the customers reside (where revenues are earned) would automatically get its share of the tax revenues.

2 Henceforth, f will refer to frequency, or the number of times a word or comment is mentioned by all the commenters. In other words, f = number of commenters) × (number of utterances per commenter). Some commenters referred to a word or topic more than once.

3 In running the word frequency query, we chose the 1000 most frequent words with a minimum of three letters of the alphabet. The word cloud represents the most frequent words measured in terms of weighted percentage of their utterances. In grouping the words, we chose the exact match words only, and not stemmed words or synonyms.

4 NVivo is a software used to analyze qualitative or unstructured data. It allows users to classify, sort and arrange information, and examine relationships in the data. Researchers can identify common or distinguishing trends to cross-examine information in a multitude of ways using its search engine and query functions.

5 The size of the boxes in Figure 2 corresponds to the level of importance of each of these five categories within the comments.

6 A Pearson correlation coefficient of 0.5 is considered strong, 0.30 is considered moderate, and 0.10 is considered weak (Weinberg & Abramowitz, 2008, p. 130).

Additional information

Funding

This work was supported by Public Interest Accounting Group at the Schulich School of Business: [Grant Number 2020-01]; Social Sciences and Humanities Research Council of Canada: [Grant Number 435-2021-0784].