Skip to main content

Taxing the digital economy

How can governments ensure international digital companies pay tax in the countries where they generate their income? Heather Self explains what is being done to crack this most difficult nut.

Since the 1920s, the arm’s length principle has been a cornerstone of transfer pricing rules, but it is not well-suited to the digital economy. Increasingly, global businesses operate as an integrated entity, with value being driven by intangible assets such as names, trademarks or algorithms.  Intangible assets are often held in low-tax jurisdictions, leaving the “market jurisdictions” (where sales are made) feeling that they are not receiving their “fair share” of taxation.

In particular, a country does not generally have the right to tax sales made in their jurisdiction, unless the entity has a permanent establishment (PE) there – a “fixed place of business, through which the business of the enterprise is wholly or partly carried on”. If significant sales are made from an overseas website, with no local PE, the market jurisdiction finds it difficult to levy a tax charge on the income generated.


The OECD/G20 Inclusive Framework on BEPS has released two reports on the taxation of the digital economy, and invited comments. The public consultation runs until 14 December 2020 and will be discussed in meetings to be held (virtually) in mid-January 2021.

The OECD BEPS (Base Erosion and Profit-Shifting) project has been running since 2013. While there has been significant progress in some areas, such as interest deductibility and the implementation of anti-hybrid measures, the challenges of the digital economy are still some way from a solution.

Following an interim report in March 2018, the Inclusive Framework (which includes a total of 137 countries, including OECD member states and a wide range of other countries) has put forward a proposal for two key pillars, which could form the basis for a long-term solution.

Even after agreement has been reached, there will be a further period before the rules come into force – 2025 is probably a realistic estimate.

Key proposals

The reports set out two key proposals to address these problems: pillar 1 and pillar 2.

Pillar 1

Companies with a large market presence may have to pay some tax in a market jurisdiction, if they have sufficient “nexus” (connection) there. This amount of tax is labelled: amount A.

In addition, companies which already have a PE or subsidiary may have to pay a standard amount of tax on local marketing and distribution activities (amount B).

Pillar 1 allocates additional tax to market jurisdictions, at the expense of the headquarters or manufacturing location.

Pillar 2

This seeks to ensure that all companies pay a minimum amount of tax, by requiring companies with a low effective tax rate to include additional income in their home country tax base; there are also provisions which would allow source countries to impose withholding taxes on amounts which only pay low taxes in the country of receipt.

How to calculate amounts A and B?

Pillar 1 requires relevant groups to carry out a complex series of calculations to determine the tax due. The questions in the consultation document cover three areas:

  • Which activities should be within scope? There are two main categories – automated digital services and consumer-facing businesses; the consultation asks what should be included in these categories.
  • What should the threshold be? The overall threshold will be global turnover of €750m (in line with the requirement for country-by-country reporting) but should there also be a threshold for the amount of local turnover?
  • What factors should be taken into account in determining whether there is sufficient nexus?

There are also a number of detailed questions on how the calculation should be carried out, including questions on how revenue should be allocated to each country (for example, how do you determine where a customer is if they are using a VPN?) and the development of a loss carry-forward regime.

Double taxation problem

A process is also needed to ensure that any double taxation is eliminated. For amount B, the questions are around what should be included in “baseline marketing and distribution activities” – should this be a narrow, or broader, definition?

A key aim of amount B is to reduce transfer pricing disputes, by providing a standard return to certain routine activities. It is likely that the calculation of amount A will give rise to disputes, so there are proposals for new dispute resolution processes.

Political hurdles

There are many political difficulties in achieving agreement on Pillar 1. The support of the US will be crucial. As most large digital companies are headquartered in the US, with significant sales elsewhere, the US may fear that it will lose tax revenues under these proposals. However, if consensus is not reached, the alternative may be worse – many countries, including the UK, are implementing their own digital services tax, which is starting to lead to more complexity and potential multiple taxation.

Pillar 2

This proposal is further advanced and probably easier to implement, although there are still a number of practical and technical issues to be resolved.

Pillar 2 builds on the concept of the controlled foreign company (CFC) regimes but seeks to implement consistent principles across the Inclusive Framework countries. However, where CFC regimes are based on a single country’s tax system, Pillar 2 would calculate tax due by reference to accounting profits, with a very limited number of adjustments. This could give rise to some surprising results, particularly if for commercial reasons there is a non-tax adjustment to accounting profits such as a revaluation of properties.

Way forward

Overall, the impact of these proposals on a global basis is expected to be a reallocation of about $100bn under Pillar 1, and potential additional tax revenues of $100bn per year under Pillar 2. However, $100bn represents only about 4% of global tax revenues, and corporate taxes are typically only about 10% of total taxes.

Applied to the UK, this would mean additional tax revenues under Pillar 2 of about £2bn a year, with no estimate yet of whether Pillar 1 would result in an increase or decrease in UK tax revenues.  The sums at stake are far from being a solution to major global tax problems, and are dwarfed by the costs already incurred in dealing with COVID-19.

This article first appeared in AccountingWEB:

Check the impact of the Spring Budget Statement with our Tax Calculator Visit our Spring Budget Hub