On February 13 the OECD issued a public consultation document requesting input on two pillars developed to address concerns about taxing the digital economy.(1) Pillar 1 deals with profit attributable to the jurisdiction of the customer or user (the market jurisdiction), while pillar 2 addresses base erosion and profit shifting.
Three options are proposed under pillar 1: the user participation proposal, the marketing intangibles proposal, and the significant economic presence proposal.(2)
In July 2015 then-House Ways and Means Committee members Richard E. Neal, D-Mass., and Charles W. Boustany Jr. released a discussion draft of the Innovation Promotion Act of 2015 (IPA 2015), designed to encourage the generation of intangible income in the United States.(3) A key component of the proposed legislation was to define intangible income to which a tax deduction would apply — so-called innovation box income and thus produce a lower effective tax rate on intangible income. IPA 2015 defined the profit attributable to non-marketing intangible income as the ratio of five-year cumulative research and development expenses to five-year cumulative total operating expenses.(4) The proposal reflected concepts similar to the BEPS action 5 definition of non-harmful tax incentive regimes for intellectual property.
This article tries to match the pillar 1 discussion with the guiding principles previously established under the OECD’s BEPS project. It introduces an expanded version of IPA 2015 as a potential way to reach consensus on the allocation of marketing income under pillar 1 and to address concerns that some local tax laws may constitute export subsidies under the WTO rules.
BEPS Defines Intangible Income
In September 2013 the G-20 members endorsed a BEPS action plan. This coordinated effort was initiated to address perceived tax avoidance resulting from aggressive transfer pricing, “to improve the coherence of international tax rules, and to ensure a more transparent tax environment.”(5) As a 2017 OECD background brief explains:
The BEPS package consists of reports on 15 actions, and sets out a variety of measures ranging from new minimum standards, the revision of existing standards, as well as common approaches which will facilitate the convergence of national practices, and guidance drawing on best practices.(6)
Final BEPS reports were issued in 2015. The purpose behind the BEPS actions 8-10 final report is to provide guidance in applying the OECD’s vision of the arm’s-length concept. The scope of BEPS actions 8-10 covers transactional profit splits; transactions involving intangibles, services, and cost-sharing agreements; and related-party transactions that, in the eyes of the BEPS working parties, “are not commercially rational for individual enterprises.”(7)
When evaluating entitlement to intangible profits within a related-party group, “members of the [multinational enterprise] group are to be compensated based on the value they create through functions performed, assets used and risks assumed in the development, enhancement, maintenance, protection and exploitation of intangibles.”(8)
The intangible profit analysis follows a six-step framework:
- Identify the intangibles and economically significant risks associated with the development, enhancement, maintenance, protection, and exploitation (DEMPE) of those
- Delineate the actual controlled transactions related to the DEMPE of the intangibles.
- Identify the full contractual arrangements and determine legal
- Perform a detailed functional analysis to identify the parties performing functions, using assets, and managing risk related to DEMPE.
- Confirm the consistency between the terms of the relevant contractual arrangements and the conduct of the parties.
- When possible, determine arm’s-length prices for these transactions consistent with each party’s contributions (9)
BEPS Action 5 — Guidelines For IP Tax Incentives
Assigning profit through a related-party contract without functional substance is a central theme addressed in BEPS action 5:
As a matter of business practice, unlimited outsourcing to unrelated parties should not provide many opportunities for taxpayers to receive benefits without themselves engaging in substantial activities because, while a company may outsource the full spectrum of R&D activities to a related party, the same is typically not true of an unrelated party. Since the vast majority of the value of an IP asset rests in both the R&D undertaken to create it and the information necessary to undertake such R&D, it is unlikely that a company will outsource the fundamental value-creating activities to an unrelated party, regardless of where that unrelated party is located.(10)
BEPS action 5 addresses profit attribution to both IP regimes and non-IP regimes and acknowledges that preferential IP tax regimes are allowed under the OECD guidelines if substantial activity related to “core income-generating activities” is aligned.
The BEPS action 5 report adopts a nexus approach, consistent with the notion that profits associated with IP activity that exceed routine profit returns inure to the geographic location where the substantive economic drivers of the IP development occur.(11) The fundamental principle underlying the nexus approach is that income should benefit from an IP regime only to the extent that the taxpayer itself incurred the R&D expenditures that contributed to that IP.(12) The report acknowledges that the proportion of functional expenditures in an entity can be a proxy for substantive activity:
The nexus approach allows a regime to provide for a preferential rate on IP- related income to the extent it was generated by qualifying expenditures. The purpose of the nexus approach is to grant benefits only to income that arises from IP where the actual R&D activity was undertaken by the taxpayer itself. This goal is achieved by defining “qualifying expenditures” in such a way that they effectively prevent mere capital contribution or expenditures for substantial R&D activity by parties other than the taxpayer from qualifying the subsequent income for benefits under an IP regime.(13)
Formulated, the nexus ratio is simply:
The BEPS action 5 report concedes that tracking IP income with IP assets may be unrealistic in some cases. It explains that in those instances:
Jurisdictions may also choose to allow the application of the nexus approach so that the nexus can be between expenditures, products arising from IP assets, and income. Such an approach would require taxpayers to include all qualifying expenditures linked to the development of all IP assets that contributed to the product in “qualifying expenditures” and to include all overall expenditures linked to the development of all IP assets that contributed to the product in “overall expenditures.” This aggregate ratio would then be applied to overall income from the product that was directly linked to all the underlying IP assets. This approach would be consistent with the nexus approach in cases where multiple IP assets are incorporated into one product.(14)
That paragraph reveals the parallels to an IPA 2015 incentive approach under which R&D incurred by the U.S. taxpayer would be the numerator of the apportionment ratio that defines IP income.
Profit Split Concepts Reflected In Pillar 1
As noted earlier, the February OECD inclusive framework report identified two tracks from the 2015 BEPS action 1 report (15) and the OECD task force on the digital economy(16):
- pillar 1, which focuses on the allocation of taxing rights and seeks to undertake a coherent and concurrent review of the profit allocation and nexus rules; and
- pillar 2, which focuses on the remaining BEPS issues and seeks to develop rules that would give jurisdictions a right to “tax back” when other jurisdictions have not exercised their primary taxing rights or the payment is otherwise subject to low levels of effective
In a May 31 roadmap document, the OECD observes that although the issues under pillars 1 and 2 are distinct, “they intersect and a solution that seeks to address them both could have a mutually reinforcing effect.”(17) The goal is to develop a consensus approach among the 129 inclusive members by 2020.
The roadmap indicates that work over the next year will include the development of rules to govern how total profits are calculated in applying the modified residual profit-split method. That method involves four steps:
- Determine the total profit to be
- Remove routine
- Determine the portion of the non-routine profit within the scope of the “new” taxing right — that is, the reallocated taxing
- Allocate that in-scope, non-routine profit to the relevant market jurisdictions through an allocation
Steps 2 and 3 can be accomplished using transfer pricing rules or simplified conventions.(18) The new profit allocation rules will effectively apply to both profits and losses.
The roadmap notes that implementing any of the proposed approaches — the user participation proposal, the marketing intangibles proposal, or the significant economic presence proposal — would likely require that specific data be available to the MNE group and the taxpayer as well as to the tax administrations. That information would include total profit, total profit per business line, sales, and users. The roadmap suggests that a potential system for reporting and disseminating the information could be based on the existing technology and framework for country-by- country reporting under BEPS action 13.(19)
Pillar 1 concepts argue for a global approach to determine income allocable to marketing intangibles, and they call for a “unifying” approach. The concept of apportioning residual profit (after routine returns) is present in each of the pillar 1 options. Further, the marketing intangibles option strongly suggests that residual profit be further reduced for trade/R&D intangible income in determining the portion of the in-scope non-routine profit. Thus, the “residual” becomes total profit minus routine profit minus trade/R&D intangible income. That is consistent with the object of aligning taxable profits with value creation. The appendix summarizes residual profit allocation concepts in each of the pillar 1 options (see paragraph 126.96.36.199 in particular).
IPA 2015 — A Prescient Global Solution?
U.S. legislative proposals emerged in 2013 that introduced tax rate reductions for profits associated with U.S.-developed economic IP.20 IPA 2015 captured the key elements of these proposals in a discussion draft, proposing a deduction on innovation box profit. Innovation box profit was derived by multiplying gross margin, minus allocable deductions, by the ratio of five-year section 174 R&D over five-year total operating expense (excluding cost of sales, interest expense, and taxes). Gross income included both domestic and foreign sales of products and services with embedded intangibles. Intangible income from the disposition of personal property, provision of services, and licensing or disposition of intangibles would have qualified.
CbC Report Provides A Starting Point
The BEPS action 13 report on transfer pricing documentation and CbC reporting provides a template for MNEs to report annually and for each tax jurisdiction in which they do business.23 The template for the CbC report asks for pretax revenue and profit by jurisdiction. The template does not require a breakdown of cost categories to arrive at pretax profit. Expanding the CbC report requirements to include revenue, cost of sales, R&D, marketing and sales, and general and administrative expense (typically separate line items in published financial statements) could provide the necessary information using an IPA 2015-type approach to allocate global pretax profit by jurisdiction and by routine, trade IP, and marketing categories (see paragraph 188.8.131.52 of the appendix).
This article’s primary intent is to emphasize an approach to define what shouldn’t be profit allocated to marketing and sales. By default, the approach would define global profit associated with marketing and sales intangibles. Having defined by default what constitutes marketing and sales profit, the question of appropriate apportionment keys to apply to the marketing profit pool becomes relevant. Obvious apportionment keys would include sales, marketing and sales headcount, marketing and sales spending by location, and digital users. The appropriate apportionment key out of the alternatives would differ by industry.
A significant piece of the non-marketing pool of profit for technology companies is global profit allocable to trade IP. If an IPA 2015 type approach were adopted to implement the OECD pillar 1 effort, it would also serve as a starting framework toward consensus on how to allocate trade IP profit geographically, recognizing that it would be necessary to weight country R&D expenditures based on DEMPE functions performed in the country. For example, trade IP strategy, financing, budgeting, and resource development performed at a headquarters location would carry a much greater weight than contract R&D performed in another country under direction from the headquarters location.
A few additional points:
- An IPA 2015 approach belies arguments (through a logical profit assignment process) that extraordinary returns should always go to
- A consensus apportionment model should minimize pillar 2 concerns because of the greater transparency and alignment with the geographic profit definition based on existing BEPS concepts. A particular country’s choice to tax at a lower rate profits assigned to that country (based on a consensus approach in assigning profit to that country) should not be overridden by another country’s view that the country hasn’t exercised its primary taxing rights. “Tax-back” rights, including global intangible low-taxed income inclusions, become unnecessary with a consensus apportionment approach based on the business functions and profit
A consensus apportionment model creates greater transfer pricing alignment and should obviate the need for Subpart F and GILTI-type inclusions.
Reprinted from Tax Notes Federal, July 29, 2019, p. 687
William C. Barrett Author