The international tax rules have undergone significant changes in recent years through the Base Erosion and Profit Shifting (BEPS) project under the aegis of the Organization for Economic Cooperation and Development (OECD). Further change is inevitable to adapt the international tax system to a rapidly digitalizing world.
The object of this article is to discuss various concepts associated with Action 1―Addressing the Tax Challenges of the Digital Economy―of the BEPS Action Plan to establish common ground between jurisdictions, and raise potential additional ideas and risks as members of the Inclusive Framework strive to reach a consensus-based solution.
The issues related to the BEPS Action 1 are numerous, and this paper only addresses a few considerations. This article also illustrates the quagmire being faced by the members of the Inclusive Framework, and suggests that trying to resolve complex issues in haste might not be in the interest of most member states and multinational enterprises (i.e., no deal in 2020 might be better than a bad deal).
The work of the OECD must be commended as the organization is under intense pressure to deliver a solution that is suitable for the economy of the 21st century. With that being said, after seven years since the initiation of Action 1 of the BEPS Action Plan, the release of the recent unified approach by the OECD Secretariat does suggest that a consensus around potential changes to nexus rules and profit allocation, and underlying concepts, still remains a highly ambitious objective.
While very difficult to achieve, and likely to give rise to uncertainty and disputes, it is nonetheless the most viable approach when considering the alternative outcome involving unilateral action by numerous states that would conceivably rely on different principles and yield limited opportunity for resolution of disputes and avoidance of double taxation.
Perspective on the Digital Economy
As the debate about the digitalization of the economy is becoming more prominent, it is important to provide some perspective with respect to the scale of the current economic activity associated with the issue. It appears that seven companies, those that are “giants of the digital economy,” fall within the most recent Fortune 500 Global Ranking (by revenue) published in 2019: Apple (#11), Amazon (#13), Alphabet (parent of Google, #37), Microsoft (#60), JD.com (#139), Facebook (#184), and Tencent (#237). Other companies notably directly impacted by the rules under design include companies such as Airbnb, Booking.com, Criteo, eBay, Expedia, Netflix, Twitter, and Uber.
While these companies and others of their like are becoming increasingly significant in their portion of the overall economy, clearly we have not yet reached the point where they represent such a large share that warrants haste in changing the rules.
However, the Fortune 500 list mentioned above also illustrates that some companies that predate the advent of the “.com” era, such as Apple and Microsoft, are included in the list. Like them, it is conceivable that many companies, if not all, will see their business model increasingly digitalized and will join the pool of companies that will eventually be impacted by the outcome of Pillar One (assuming there is some form of agreement).
Pillar One Unified Approach
In October 2019, the OECD Secretariat released a consultation document detailing the outline of what it has dubbed the “unified approach,” which brings together the “without prejudice” proposals put forth by the delegates of the Task Force on the Digital Economy, namely: (1) the “user participation” proposal spearheaded by certain European countries, (2) the “marketing intangibles” proposal spearheaded by the U.S., and (3) the “significant economic presence” proposal spearheaded by the G24 and in particular India.
While the OECD Secretariat has made a commendable effort to bring together the seemingly different proposals, it remains that conceptually, the unified approach is somewhat of a misnomer, given that it’s not clear that all proposals are trying to solve the same problem (perhaps “aggregated approach” would be a better reflection of the proposal).
While the three proposals do share some commonalities, their perspective on the issue arising as a consequence of the digitalization of the economy and the manner in which to address the consequent tax challenges are very different. For example, the user participation approach really targets internet-based companies/business models that have emerged in the last two decades, while the other two proposals are much broader in scope in terms of the types of companies/business models being targeted. Putting aside the technology companies that are more likely to operate fully without physical presence, and considering traditional businesses that further digitalize, it is not evident that the latter will entirely forego physical presence, and therefore it’s not clear that a new set of rules around nexus is warranted for traditional business models.
Moreover, the framework within which the approaches try to achieve results are different. The marketing intangibles approach appears to aim to change profit allocation rules within the arm’s-length principle, while the other two approaches seem willing to depart from it.
Despite the best intentions of the OECD, it does not appear that the Secretariat, in its efforts to date, has been able to bring together the ideas into a framework that is fully cohesive. The unified approach is a construction put forth by the OECD as a proposal for negotiation by its constituency. Perhaps the principles achieved under consensus will be better integrated and sufficiently robust to withstand the test of time. Absent such an outcome, it’s unclear whether the unified approach is a theoretically sound approach to forge the new rules needed to discourage unilateral tax measures.
Digital Economy vs. Highly Digitalized Businesses vs. the Economy in 2050
It is notable that the BEPS Action 1 Final Report published in 2015 was aimed at the “digital economy” while the BEPS Interim Report published by the OECD in 2018 shifted the lexicon, referring to “highly digitalized businesses.” While characteristics of highly digitalized businesses (i.e., scale without mass, reliance on intangibles, data and user participation) are provided, there is no clear or definitive definition of what may be considered a highly digitalized business. It appears that the implicit definition the OECD is putting forth is rather a deductive process whereby highly digitalized businesses can be defined by what they are not, namely the extractive industry and other producers and sellers of raw materials and commodities, the airline and shipping industry, and the financial services sector (mostly).
This absence of clarity is problematic because it can mean different things to relevant stakeholders, and it also fails to address the dynamic shift that is pervasive in the economy as members of the Inclusive Framework try to revise the rules. Given the importance of intangibles and data/user participation, there is significant risk that some people are currently failing to foresee that many industries that might be thought of as “non highly digitalized” are actually very likely to become highly digitalized businesses in the not too distant future. Examples include the appliances industry providing smart refrigerators, the agricultural industry using dynamic weather and soil data, and the engineering industry using virtual reality and other tools to deliver services remotely. In other words, we very soon will be faced with the situation that the challenges of taxing Google or Netflix might be more similar to the challenges of taxing Whirlpool and Syngenta than one might think, and it’s not clear that the consumer-facing businesses concept will appropriately address the problem.
Ultimately, it appears that the design of the OECD solution should be based on a consensus around what problem(s) is being identified as in need of a solution. Is the goal to address issues related solely to internet giants or is the goal to come up with holistic new rules for the evolving business models across numerous industries that are leveraging digital technologies and that are significantly relying on market/marketing intangibles? Ring-fencing any portion of the economy appears very problematic, as it is bound to require constant remodeling and will potentially provide for non-neutral tax policies. If the rules are clear and well-designed, then there should be no need to ring-fence any portion of the economy, and if one takes a long-term view on the matter, this is most likely the best approach.
Value Creation in the 21st Century
The concept of value creation is paramount in international tax when it comes to transfer pricing. The OECD BEPS Actions 8-10 Final Reports published in 2015 were named Aligning Transfer Pricing Outcomes with Value Creation. Interestingly, there is no definition in the latest version of OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administration (OECD Transfer Pricing Guidelines), which were published in 2017, with respect to the meaning of “value creation.”
The OECD BEPS Actions 8-10 Final Reports do give an indicia of the intended meaning: “It is important to understand how value is generated by the group as a whole, the interdependencies of the functions performed by the associated enterprises with the rest of the group, and the contribution the associated enterprises make to value creation.” Also, it was recognized that the exploitation of intangibles can account for either a large or a small part of the MNE’s value creation.
In the 20th century, the link between value creation and nexus was generally clearer. For example, if you were to extract oil from a well, it was unlikely that you would consider not assigning profits to the jurisdiction that had the oil reserves. Back in May of 2017, an Economist article was titled “The World’s most Valuable Resource is no Longer Oil, but Data.” In the 21st century economy, data is expected to be extremely valuable. The issue we face is that currently you can “extract and exploit” digital data without nexus, and as a result some companies have sometimes opted to assign no profit to the market jurisdiction.
With this assessment in mind, a few observations can be made. First of all, value creation might have become a concept that needs to be revisited to align with the 21st century zeitgeist. What seems to be ongoing is that many countries perceive a flaw in the value creation concept since it can lead to no profits left in the market jurisdictions where, for example, services are consumed, advertising is viewed, and user-participation occurs. Perhaps the international tax jargon should include a new concept named “value crystallization,” which would help to bridge the gap between nexus for physical resources and nexus for digital resources (namely, data). The concept of value crystallization is closely intertwined with the concept of “market jurisdiction” that the OECD has come up with in the Interim Report.
Of course, governments do not grant rights to mine digital data like they do for natural resources, and digital data mining can be done remotely. For reasons beyond tax, such as privacy concerns, governments have established that personal data of users remains the property of those users. However, for tax purposes only, perhaps data could be deemed to be an asset owned by companies, and governments could potentially force companies to perform digital data collection locally through a local legal entity. This would allow countries currently implementing digital services taxes a path to revert to conventional income tax given that the local subsidiary would own the data intangibles that are necessary to create value.
The concept of value crystallization could be helpful for business models involving the provision of remote services and data streaming (e.g., e-commerce, travel/lodging industry, online advertising, content streaming). Clearly, no highly digitalized company can create value without actual consumers/users.
Finally, it is worth noting that Pillar One could potentially be seen as incongruent with consensus reached by member states with respect to BEPS Actions 8-10, which primarily serves to reward substance and attribute profits to people functions. Indeed, Pillar One, through amount A, assigns profit where there are little or no people functions or control and management of risks. Hence, the logical step to bring the equation into equilibrium is to posit that Pillar One’s missing variable is assets and that amount A is really a remuneration for intangibles associated with the market jurisdiction. To this end, the current discussion on amount A does little to reconcile relative returns to the overall assets or even cumulative intangible assets of the MNE.
Intangibles and Profit Allocation
In the context of transfer pricing, the issue of intangibles is a challenging one. The new version of the OECD Transfer Pricing Guidelines published in 2017 states that an intangible is something “which is not a physical asset or a financial asset, which is capable of being owned or controlled for use in commercial activities, and whose use or transfer would be compensated had it occurred in a transaction between independent parties in comparable circumstances.” While refusing to provide a holistic list of intangibles, the OECD Transfer Pricing Guidelines do provide two broad categories of intangibles, namely trade intangibles and marketing intangibles, that encompass a significant umbrella of intangibles.
The definition of “marketing intangibles” found in the OECD Transfer Pricing Guidelines reads as follows:
“An intangible (within the meaning of paragraph 6.6) that relates to marketing activities, aids in commercial exploitation of a product or service and/or has an important promotional value for the product concerned. Depending on the context, marketing intangibles may include, for example, trademarks, trade names, customer lists, customer relationships, and proprietary market and customer data that is used or aids in marketing and selling goods or services to customers.”
Interestingly, the revised definition does not make any mention of “consumer data.” Consumer data is the trail of information left behind, mostly in digital format, by people through daily interactions, and this includes information left behind while using various websites, platforms, and technological applications. Consumer data should not be conceived as overlapping with “customer relationships,” “proprietary market data,” and “customer data.” Consumer data has become one of the key value drivers for some highly digitalized companies that try to monetize the vast amounts of digital data that they gather from their users on a daily basis.
Perhaps consumer data is not a marketing intangible, but given the definition found in the OECD Transfer Pricing Guidelines, it is surely an intangible, and it should be recognized as such in future revisions to the guidance.
Taking the premise that the market jurisdiction should earn the profit associated with the consumer data (which I believe is a reasonable viewpoint), the issue becomes how to determine the value of the data. Within the arm’s-length principle framework, one of the potential avenues to assign value to consumer data is to look at the profits earned by companies that manage loyalty programs that is not specific to one retailer (e.g., Air Miles, Maximiles, PAYBACK). Given the overall nature of this type of business, it could provide a reasonable estimate to allocate a portion of residual profits to market jurisdictions. Like comparable distributors are used to assign an operating margin for a limited risk distributor, the range of operating margins/returns earned by loyalty programs could be used to attribute an operating margin/return on locally derived revenues of specific digital companies to reflect the contribution associated with consumer data.
In thinking about the allocation of residual profit, it must be remembered that some industries/businesses that are coming to life, and whose value proposition is based on digital technologies, are naturally monopolistic (e.g., Facebook, Google). It appears troublesome to design tax rules without isolating the challenges of taxing highly digitalized businesses from the public perception associated with very high profit margins due to competition issues. After all, the discussion draft seems to primarily address the circumstances in which highly digitalized businesses give rise to significant residual profits. The real challenge arises when there is no residual profit or insufficient residual profit to compensate all of the jurisdictions. Considering the potential threshold of EUR750 million in revenues, matters become even more complicated if we take into account intertemporal effects—namely, that significant losses (or investment) may have been borne by the core IP owner in earlier years and long before reaching the revenue threshold. Members of the Inclusive Framework should bear that in mind as they continue the negotiations and public consultations.
Location of Activity vs. Market Jurisdiction
Finally, in the 21st century virtual world, determining the location of consumers and targeted participants is becoming blurry and needs some critical thinking to refine the market jurisdiction concept.
If I decide to purchase a good from my smart phone while on vacation in Europe for next-day delivery at my hotel while being billed at my regular home address, is the market jurisdiction my home country or my vacation host country for this purchase? If we determine that it is the host country, will the companies be able to carve out the transactions for purposes of profit allocation and reporting? What happens if I decide to have the merchandise delivered to my home while the server localization will have identified me as being in Europe?
If a Japanese person planning a trip to New York reviews Yelp! comments written by fellow countrymen in Japanese character about the best Japanese restaurants in New York to make a reservation before the trip, is the market jurisdiction Japan or the U.S.?
While the issue’s relative importance is not paramount, some regions (e.g., Europe) with great mobility and proximity to other jurisdictions may find this issue more significant.
The tax challenges associated with the digitalization of the economy are very significant. While it appears that a consensus has arisen around the need to change nexus rules, many other critical issues remain unresolved. While many interesting concepts have been developed, further ideas and thinking outside the box will be necessary to try to achieve consensus.
Changing the international taxation rules is the right thing to do. The real challenge is doing things right, and an expedited process is very unlikely to result in a robust long-term solution. Only a thorough brainstorming will achieve a long-term suitable solution for the members of the Inclusive Framework.
This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.
Alexandre Mercier is a senior manager in the transfer pricing practice of PricewaterhouseCoopers in Montreal. All views, thoughts, and opinions expressed in this article belong solely to the author, and not to any entity with which he is, has been, or will be associated.