In my previous contribution to the Cayman Financial Review, I discussed the European tax agenda for 2019, focusing on the EU initiatives in respect to the taxation of digital economy, as well as regulatory measures adopted by individual member states. Not only will the introduction of a tax on revenues from digital economic activity likely have a negative impact on competitiveness, it also reflects a highly aggressive approach to changing the regulatory framework of international taxation. Specifically, the advocacy of a digital tax is not rooted on the notion there is a need to (further) curb tax evasion (or avoidance) of Amazon & Co in the spirit of the BEPS reform, but it is rather built on the idea that there is a part of economic activity which currently is untaxed, which is perceived as inherently unjust by advocates of the digital tax.
As explained in the earlier contributions, the notion of untaxed economic activity is based on a fundamentally new concept of the ‘value of things’; i.e., the idea that value primarily depends on how and where people use specific ‘things’. Due to the glaring vagueness of the underlying idea, there is no consensus on how to translate the respective notion into sensible tax policy. In the given situation, I suggested that regulation based on ‘soft law’ was preferable to ‘hard law’ alternatives. While I do stand by my earlier arguments, the developments of last months have spawned doubts about how to best facilitate the adoption of such soft law regulation.
Having been of the opinion that the OECD, at least in the realm of transfer pricing, has generally been doing an adequate job of sustaining a coherent international framework (i.e., notably by maintaining consensus on the arm’s length standard), I advocated to rely on the OECD, in its role standard-setter, for taking the lead on shaping the regulation on digital taxation. I phrased this as follows: “Many neoliberals have a hard time accepting that the OECD can actually play a positive role. It has been said that BEPS is a concerted effort by high tax countries to sustain (and expand) their tax bases. There is certainly merit to this assessment. The pragmatic question in this context, however, seems to be whether accepting the OECD as a facilitator of soft law is the lesser of two … evils.”
The rationale for advertising a central role for the OECD was that sustaining consensus on the arm’s length standard as a sort of ‘meta regulation’ is sensible. In a nutshell, tax competition is preserved, and taxpayers can trust that national tax authorities will base their assessments on a coherent overarching principle. Looking at the position adopted by the OECD in the context of the public discussion on digital taxation, however, I feel the need to recant.
The OECD discussion draft on digital tax: One big disappointment
While we have seen many public discussion procedures, the public consultation document (PCD) on digital taxation is easily the most disappointing. To be sure, such policy papers are intended to facilitate discussion rather than presenting a thought-out policy proposal. In the case at hand, however, the proposal was ‘framed’ in a highly biased way. Specifically, the OECD failed to differentiate between untaxed digital activities (UDAs) and the earlier BEPS reforms. This is not a technicality, as a sensible discussion about international tax policy requires an accurate distinction of the issue at stake.
BEPS was focused on modifying the international tax regulations, including arm’s length-based transfer pricing, in a way that the resulting profit allocation between related (separate) legal entities would be (more closely) aligned with the commercial reality (or substance).
In other words, BEPS was explicitly targeted at preventing tax avoidance based on tax and transfer pricing structures that were deemed “high[ly] aggressive”; e.g., licensing structures that syphon the profits of multinational enterprises (MNEs) to entities without substance (i.e., zero or few employees) located in low-tax territories.
The “untaxed” digital activities now being discussed by the OECD, however, are not a result of aggressive tax planning by MNEs. The activities discussed are rather how Facebook (and other digital platforms) are being “used” by the consumers (“users”); i.e., the fact that there are Facebook users in a jurisdiction (“market country”) in which there is no legal entity of Facebook, as well as cases in which local entities merely render minor local support activities.
Hence, the proposals on the taxation of the digital economy have a fundamentally different aim compared to BEPS; namely redefining (broadening) the concept of what constitutes (taxable) value creation.
Whether or not the activities of local users, which from the perspective of MNEs are independent third parties (which qua definition are interacting on an arm’s length basis), should be subject to corporate taxation will be discussed in the concluding part of this article.
Such a discussion, however, must not be framed in a context suggesting that respective tax regulation is aimed at reducing tax avoidance. In other words, such proposals cannot be allowed to be promoted under the same ‘umbrella’ as anti-avoidance regulation such as BEPS. Framing the discussion in such a way, grants advocates of respective proposals a cloak of legitimacy they do not deserve. The OECD proposals contained in the public consultation document convey an overly pessimistic evaluation of the effectiveness of BEPS and can be interpreted as suggesting that a substantial erosion of the tax base is attributable to untaxed digital activities. The OECD position reflected in the public consultation document can be summarised by the following citation (PCD, Paragraph 65): “… by failing to acknowledge the reality that businesses can today have an active presence or participation in market countries without a physical presence, or one that would justify a substantial allocation of income to that jurisdiction, the existing international tax rules fail to properly allocate income to the locations in which an enterprise is understood to create value in today’s increasingly digitalised world”.
There are at least two intriguing aspects – and several related questions:
i. Is the current system really failing to allocate income “properly”? Does this notion imply that BEPS was not effective or are we really talking solely about a different concept of “value”?
ii. Now, if there is a physical presence (legal entity) within a market country, i.e., there is a taxable nexus, what is meant by “justify a sufficient allocation of income”? Would that notion not imply that the arm’s length profit allocated to such a local entity is not deemed “sufficient”?
In regard to the first aspect, it is highly unfortunate that the OECD neglected to reconciliate the policy proposals with any empirical analysis of the problem allegedly caused by untaxed digital assets (perceived tax gap?). As emphasised above, the OECD failed to delineate between UDAs and BEPS and, it seems clear, the perceived problem can only be attributed to a new concept of value creation; i.e., a concept of value creation that was alien to the BEPS.
It is the second question, however, that really is worrisome. Here, the OECD is embarking in a direction that is difficult to follow. What the OECD is essentially saying is, that the arm’s length principle (i.e., even when adhering to the post-BEPS interpretation including the DEMPE1 concept and other substantial reforms) is systematically not fit for ensuring an “appropriate” alignment of value creation and profit allocation. This is cause for concern for multiple reasons; first, “appropriate” or “fair” are weasel words that hardly constitute an acceptable basis for designing international tax regulations, and second, the OECD opens the ‘Pandora’s box’ by implying that traditional transfer pricing concepts, which are based on remunerating MNE entities performing low-value added, routine functions with a small but stable profit (e.g., a mark-up on costs), will no longer be considered viable in cases were “marketing intangibles” are assumed to exist.
These marketing intangibles (in simplified terms) can essentially be tied to “digital activity” that takes place in the market country (i.e., the mere existence of Facebook users).
Identifying such marketing intangibles will entitle the market country (the legal entity located there) to participate in the entrepreneurial profits of the MNE; i.e., there will no longer be a “remuneration” for local activities, but rather an allocation of global profits of the MNE (which the OECD much too generously ascribes to the existence of marketing intangibles). Worst of all, the OECD proposes formulary apportionment approaches (the anti-paradigm to the arm’s length principle) to calculate the share of market countries in the entrepreneurial profits.
Without discussing formulary apportionment mechanisms, it should be obvious that the effect will be a substantial shift in tax revenues to market countries, which tend to be large, high-tax economies.
It is hard to say where the OECD proposals on digital tax will ultimately lead. The OECD position, however, can be utilised by market countries as a justification for comparatively radical reforms. Market countries will have an irresistible incentive to emphasise the importance of marketing intangibles. When considering that the OECD, in parallel to the digital tax proposals, is also exploring whether the idea of “global corporate minimum tax” remains relevant, one cannot help but feel that the OECD is acting as an accomplice of a global corporate tax push rather than a principled standard-setter that facilitates consensus on the arm’s length standard as a market-based meta regulation.
Such an unprincipled position also erodes the understanding that jurisdictions are sovereign in setting their tax rates and that low tax rates alone are no cause for ostracism. Re-emphasising this understanding should be rather easy when you are confident in the BEPS reforms being effective. As there is no indication that BEPS will prove futile, I do not comprehend the promotion of hyper-aggressive tax regulation reflected in the public consultation document. Hence, I cannot, in good conscience, sustain my advocacy of a central role for the OECD in the context of digital taxation.
Do we really want to tax digital activity of users?
This is really the question that the OECD should have focused on much more diligently. Despite the critical comments made above, which are attributable to my concern about the erosion on the consensus of the arm’s length principle, there is no denying that the taxation of social media platforms, search engines and online market places is indeed difficult.
I can understand why there is a debate about the ‘nexus’ issue. What I do not understand, however, is that the nexus question is presented by the OECD as being inextricably tied to the question of profit allocation. The introduction of digital permanent establishment could, in my opinion, solve or at least mitigate the nexus issue, while application of the arm’s length principle would be suitable and feasible for profit allocation purposes. Granted, such an approach would be a more evolutionary policy reform, but what is wrong with that? To me, the entire debate is fuelled by the notion that the “users” of social media platforms create substantial value. Looking at individual users, this notion seems absurd. If you upload your kitten pictures on Facebook, that does not entitle you to any remuneration – let alone to share in the residual profits of Facebook.
You get free access to the platform and can set up your account, while in exchange Facebook gets access to your data and can display advertising to you. That is an arm’s length transaction. You do not like the arrangement, you do not set up an account. Some people make the argument that the users do not “understand” the value of their data and that Facebook profits of their ignorance. Really? Aside from the obvious paternalism, where do you draw the line? Will other market transactions also be second guessed for tax purposes?
The notion of value creation underlying the OECD proposals is that once you exceed a critical mass, these collective accounts are suddenly entitled to a share of the residual profits of Facebook and taxable by the respective market country. How such a critical mass can be defined in non-arbitrary terms is beyond me. Obviously, the customer base is of great value for Facebook, but is that not true of all customer bases, including those of traditional business models? I also should be pointed out that Facebook does not get a free ride on the infrastructure of the market country. There are no externalities that would justify compensatory tax. I would rather make the argument that Facebook will facilitate the creation of new (local) business that will in turn be taxable by the market country – i.e., there are positive externalities rather than negative ones.
Whether or not the above assumptions and questions regarding value creation are valid, is of secondary importance. What is important is to acknowledge that there are no foregone conclusions and that we need a serious debate – an opportunity wasted by the OECD. One thing we have to take seriously, however, and a lesson to be learned from BEPS, is that we need to ensure and communicate that the value-adding activities of MNEs; programming, marketing and the creation of all the respective intangibles, are taxed – and that profit allocation and taxation are aligned with the physical location where the activities are actually performed. We need to work towards building trust in utilising the post-BEPS assessment framework and a modern interpretation of the arm’s length principle. Establishing such trust will be the most effective safeguard against radical reforms and an unprincipled global corporate tax push.
To end on a bright note; the EU initiative for a digital tax was rejected in March, as Ireland and the Nordic member states remained opposed. So, at least for Europeans not being subject to the national versions of the digital tax, there is a reprieve that can be utilised to highlight that a digital tax on revenues is neither sensible for curbing tax avoidance nor for promoting any notion of “fairness”. The rejection of the digital tax could also dampen the enthusiasm for other ill-conceived centralisation initiatives such as the Common Consolidated Corporate Tax Base. Let us hope for the best. And, sadly, let us also hope the OECD does not further undertake to reinvigorate the high-tax member states in propagating centralisation and the introduction of a digital tax or a global minimum tax.
1 DEMPE is the concept of development, enhancement, maintenance, protection and exploitation of intangibles, which has resulted in significant changes in how multinational enterprises implement the arm’s length principle for transfer pricing.