Taxation of the digital economy is one of the more difficult and contentious issues in international taxation. There are several strands to this complexity. One strand is profit attribution for enterprises trading across different tax jurisdictions. Yet, several countries have introduced or propose to introduce a tax on MNEs in the digital sector known as a digital services tax (DST). The EU Competition Commissioner and possible candidate for European Commission presidency, Margrethe Vestager, has called on European countries to take the lead, if global initiatives led by the OECD do not produce a solution in a reasonable period of time.
While many EU Member States support a DST as set out by the European Commission, there is considerable opposition from some countries such as Ireland. This effectively blocks the introduction of a DST at EU level as unanimity among MS on tax matters is a requirement for Directives on Taxation. There is also difficulty in reaching agreement both within and outside the OECD. The OECD considers reaching a consensus-based solution by 2020 to be a challenging objective.
There is also opposition from some non-OECD countries. For example, a recent report by the Indian Government notes differences in the OECD approach to the allocation of profits to Permanent Establishments (PE) and the UN Model Tax Convention. A change introduced is that the OECD required that profits be attributed to a PE on the basis of functions performed, assets used and risks assumed (FAR). A second major change was the removal of the option of apportionment in determining profits. The Indian Government has argued that these changes ‘are detrimental to the interests of developing countries’ and have not been accepted by the Indian Government. The report also considers that the ‘interests of [OECD] member countries take precedence over the interets of non member countries in OECD decisions’.
Users as the Source of Profits
Several countries, for example India, have argued that users of digital services create value which in turn leads to the allocation of profits to user jurisdictions. Users are seen as being analogous to employees, or other providers to the business.
A UK Government position paper argues that ‘user participation creates ‘value for certain types of digital businesses through their engagement and participation’. The Australian Government also considers that user created data is very significant for some digital businesses, for example social media companies, but not for all digital businesses, for example on-line market places. Similarly, the EU in its proposals for a DST considers that ‘users of digital services are increasingly part of the value creation process, consciously or not’.
The OECD’s recent discussion paper sets out certain tax regimes for taxing digital profits. One proposal envisages ‘for certain businesses, an amount of profit be allocated to jurisdictions in which those businesses’ active and participatory user bases are located, irrespective of whether those businesses have a local physical presence’. The profit to be allocated would consist of ‘the residual or non-routine profit of a business’. The OECD states that ‘the value of users, and the users of each country, to a business‘ would be calculated via a formula.
An Alternative Explanation
Even though the digital sector as a whole is the focus of much analysis, there is a small number of firms that dominate digital commerce. These firms have global ambitions. For example, Google (Alphabet Form 10k) states “Google’s mission to organize the world’s information and make it universally accessible” and explains ‘How We Make Money’ as follows:
‘The goal of our advertising business is to deliver relevant ads at just the right time and to give people useful commercial information, regardless of the device they’re using. We also provide advertisers with tools that help them better attribute and measure their advertising campaigns across screens’.
Facebook (Form 10k) states ‘Our mission is to give people the power to build community and bring the world closer together’ and makes a similar statement to Google about the source of profits.
A more insightful description is given by Zuboff (The Age of Surveillance Capitalism, 2019). Zuboff argues that Google users are not customers, nor do they function as workers. Rather, they are sources of ‘raw-material supply’ that is data/information derived from all digital interactions, for example online searches, emails, contacts, travel, etc. This has resulted in behavioural data beyond that required for service improvement. Zuboff refers to this surplus data as ‘behavioural surplus’ (Zuboff, p. 75) and is the source of the market power and profit maximisation for the digital giants. This ‘behavioural surplus’ has been facilitated by anti-competitive practices and poor or non-existent regulation.
One example of the latter is Google mapping services. Google has a vision to map the world (Zuboff, pp 151-155). The buildings street scape, and land were mapped without the consent of the owners of buildings, land and other features, and who may not use mapping services. Digital data may be and is often shared across different digital enterprises. Accumulation of information and knowledge, often without the knowledge or consent of users and non-users, in conjunction with large expenditures on data storage and artificial intelligence has enabled much more accurate prediction of the effect of advertising on consumer expenditures. Such computing power may not only satisfy demand but also create demand and thus introduce an element of control with consequent reduction in uncertainty.
It is likely that data analytics and artificial intelligence will become much more central to mass market firms such as airlines, car insurance companies monitoring driver behaviour and risk, and health insurance companies.
Some Implications for Tax Policy
The rationale for a digital services tax is often given that ‘users’ create value and that this value should be allocated to users’ jurisdiction and taxed accordingly. However value added arises from both users and non-users of digital services. It is combined from different platforms under different ownership. It includes both digital and non-digital or real world sources of information such as Google mapping. Identification of users with value added is likely to become increasingly different from ‘behavioural surplus’ with consequent implications for a DST based solely on users. Rather, tax based on profit measurement and profit allocation as with the CCCTB case is likely to be more equitable and effective.