A tax challenge: how make digital players contribute?

A tax challenge: how make digital players contribute?

On 1 July 2021, the Organisation for Economic Co-operation and Development (OECD) announced that 130 of the 139 countries thinking together on international taxation (united in the Inclusive Framework) have found an agreement in principle on what is called a fundamental reform of international corporate taxation. The intention is to further develop these principles between these 130 countries by October 2021, with the aim of applying them effectively in 2023.  The European Commission has previously announced that it will present its plans for taxing digital players ('the digital levy') in July 2021 and has already said that these will be applied alongside any plans from the Inclusive Framework on corporate taxation.  However, after the endorsement of the agreement in principle by the G20 on 10 July 2021, the European Commission announced that, for the time being, it is prepared to abandon its proposals on a 'digital levy'. In this article, we explain this agreement in principle, estimate the consequences and present an alternative (and faster) route for the taxation of digital activities in Belgium.

Wanted Law asked lawyers Paul Verhaeghe and Roel Deseyn, our two guest bloggers.

On what exactly was an agreement in principle reached between 130 of the 139 countries in the Inclusive Framework?

130 countries representing 90% of the world economy have agreed in principle on the need to change international tax rules as of 2023 in such a way that countries can better tax profits obtained through digital and other activities on their territory.  This is done through two techniques, called 'pillars' in the jargon. Remarkable: among the nine countries that do not support this agreement in principle, there are three Member States of the European Union: Ireland, Hungary and Estonia.

Pillar 1

The first pillar would generate an estimated USD 100 billion more in tax revenues for countries worldwide each year. This pillar creates a mechanism that makes it possible to tax accounting profits - on a group level - from digital activities in several countries.  Currently, only the country where the tax residence of the head office of that group is located can subject the profits booked there at group level to corporation tax. According to the agreement in principle, these accounting profits from digital activities will be divided between the countries where part of the turnover is generated, using complex formulas that are yet to be determined. The groups that can thus be taxed by several countries on the group result must have a global annual turnover of at least 20 billion euros and an overall accounting profit margin that exceeds 10 %.  A maximum of 20-30% of the profit in excess of 10% generated by income in other countries may be taxed by those other countries. These countries then tax according to the country's share of the group turnover provided that a minimum annual turnover of at least €1 million is achieved through digital activities in the country in question. For smaller or economically weaker countries (these are the countries with less than 40 billion euro gross domestic product), there is a different arrangement. These countries can also claim a share of that profit at group level if at least EUR 250,000 of that group's turnover for that year was generated in their territory.

Pillar 2

The second pillar would raise USD 150 billion annually in tax revenues worldwide, through a right for countries to apply differently (see below) tax rules on payments to foreign countries as a result of digital and other activities to companies that are established in their country or have a permanent establishment there.  The condition for such different application is a lack of a minimum effective corporate tax rate of 15 % in the other country where the payment is collected on the accounting profit generated by those payments. Countries may choose not to apply the second pillar.  This is logical, after all it concerns their national corporate tax and so there is autonomy and sovereignty. Groups must have a global annual turnover of at least EUR 750 million in order to fall under these optional rules of different national treatment.  Countries can then reject booked expenses for tax purposes if the recipient of the payment is not subject to an effective corporate tax of at least 15% on the profit made from those charged expenses.  Or they may provide those payments made to companies in other countries be subject to a tax corresponding to the difference with an effective corporate tax rate of 15 %.  Thus, income paid to a group member may be subject to tax in the country of the paying group member if in the country of the receiving group member there is not a minimum effective corporate tax of 15 % on the profit arising from the payment received.

What is the problem with the digital economy on which those 130 countries were seeking an agreement?

The digital economy consists of all activities that create wealth through digital means. Given the market value of personal and market profiles and the use of online platforms, the classically taxed categories of goods and services are no longer sufficient to capture wealth in a digital economy. 

Technological innovations in recent decades have led to significant cost reductions on the customer side in terms of connecting directly to businesses, wherever they are located in the world. These technological innovations have also led to a significant increase in the type of services that such companies can offer.

Free services to obtain the data collected from users lead to an increasingly important economic value, as these data are essential for efficient publicity or powerful marketing. This new reality poses problems in determining the scope and localisation of the wealth created, in this case by the marketing of user data. 

International groups can offer the market profile of Belgian users, who on average have a higher purchasing power than users in most other countries, for sale worldwide without paying taxes in Belgium.

Whenever new modes of production allowed for new forms of wealth creation - and thus altered the balance of the ability to pay between taxpayers - new types of taxes were devised to achieve a fairer balance according to the altered power between taxpayers.

The invention of steam engines meant that large sections of land were no longer needed to produce a lot, so that in time taxes were no longer based solely on land ownership but also on the number of goods produced (the patents). The invention of the exploding engine reduced production costs with more value being created by movable property, so that in time it was decided to tax on income and profits rather than on the externalities of property ownership.

Impact of digital activities

Digital activities today lead to new forms of wealth creation, which are accounted for in various ways through abstract valuation (value). These digital activities have specific characteristics: they often allow high profit margins, no longer require a physical presence in the country where the turnover is generated and allow better tax avoidance. Tax avoidance via digital activities is of course also possible for companies with a physical presence, for example by locating software licences and servers abroad and thus shifting part of the profit made in Belgium (by selling goods and offering services to companies and consumers) to another country.

Digital activities therefore lead to a change in the balance of the ability to pay  between taxpayers and, as a result, to political tensions over what constitutes fair contributions for these new forms of wealth. The average difference in tax burden between non-digital business models (23.2%) and digital business models (9.5%) was calculated by the European Commission for the justification of a draft directive on taxation of digital activities in March 2018. This was therefore already about 14 % difference in tax burden at that time.

Not being able to count on contributions to urgent budgetary needs according to this changed ability to pay is internationally regarded as unfair. It is therefore logical that new forms of taxation are being sought on many fronts as an adequate response to these recent phenomena of unequal contributions.

Several large digital players - such as Amazon - prefer not to make an accounting profit on the goods they sell in Europe in order to acquire larger shares of the European market, because less profit means less tax and thus offers a competitive advantage.  For example, on European sales of tens of billions of euros (consisting of the delivery of goods), Amazon reports an accounting loss or a minimal profit. By selling the data of their European customers for marketing purposes or through the intra-group charging of copyrights for the use of the platform, it is possible to make a substantial profit at group level from these seemingly loss-making European activities.  Of course, this is only possible with a fiscal optimisation whereby eager use is made of fiscally favourable regimes in certain countries outside and within Europe.

We have to conclude that in the agreement in principle, the relationship between the countries is still thoroughly skewed and does not adequately remedy such excesses of a digital business model.  Only countries with 'buyers' of customer preference data have attributable revenue, while the wealth that is sold actually originates in the countries where the buyers enter their preferences into Amazon's platform and that data is 'harvested' (read: gathered) there.  However, accounting profit does not allow mesuring that harvesting of data because it does not generate an accounting profit that can be attributed to a share of group profit.  That is precisely why the OECD is looking to balance things out with complex formulas. We shall see what is decided on this matter by October 2021.  Until an agreement is found on the implementation, this pillar 1 is a giant with feet of clay.

Globalisation, digitalisation and intangible goods characterise the digital economy and oblige to rethink international corporate taxation from the point of view of taxing the place of collection of the data that is later on sold, rather than the place of the buyer or seller of that data. In the future, a more prominent role in the distribution of tax revenues will certainly be played by the country where services or goods are actually ordered, delivered or used. Turnover from the classically used concept of sales leading to accounting profit thus no longer includes the right place where the real 'new wealth' (data collection) is created and can be sold afterwards.  We are thus alluding to the place where the data is harvested. In the digital economy, this data will subsequently be offered to the highest bidder worldwide.

Only a technique that allows for taxation at the place of data harvesting on the basis of the turnover that is achieved there and elsewhere thanks to this data at group level and not on the basis of a mere accounting profit at group level can counteract deloyal tax competition in a digital economy.   It concerns the fundamental difference between the creation of wealth and the accounting rules that try to express an arbitrary value for it. By basing their approach on profit, the 130 countries that have signed up to the agreement of principle are working with the wrong starting point.

No breakdown of physical criteria of business assets can be made to link a profit made to digital activities in a country such as Belgium, so that one has to fall back on the own accounting and tax definitions of the country where the company is located.  Similar digital activities in Belgium are then taxed differently due to differences and variations in accounting concepts of profit, depending on the legal or administrative definition of profit in the different countries of residence. 

Group profit can also be depressed by major investment decisions in a country. Think of servers or distribution centres in large countries.  This reduces the cake to be shared between all countries.  A country like Belgium with a lot of purchasing power will then have less tax revenue, although a multinational group will generate a lot of turnover here. This is only possible because large costs are incurred elsewhere that only benefit the country where the investments are made.   This is another major undesirable effect of a system that does not abandon accounting profits to tax digital activities. Effective taxation based on accounting profits is therefore not efficient in achieving an equitable distribution of the right to tax wealth created by digital assets.

In the technical approach of the 130 countries, neither equal treatment with Belgian companies nor a level playing field between competing companies with digital activities in Belgium is possible as long as there is no universal system of accounting standards. This problem of definition is exacerbated by the fact that it is very difficult to write off the value of the digital activity or the clientele according to traditional accounting principles, since the value of a digital activity can disappear overnight, for example, due to a website of a competitor that users suddenly choose en masse as their preferred tool. Fair and efficient taxation therefore requires that the accounting concept of profit be abandoned for taxable digital activities and that a taxable base based on turnover be established in the country where the data are harvested. In our proposal, the taxable base is determined on the basis of legal presumptions that allocate a portion of the turnover to Belgium according to the effects of that digital activity.

The fact that some people do not contribute or contribute only minimally, or certainly not in proportion to their real ability to pay, is even more unfair in a context where, because of the coronavirus, climate change and the ageing population, a massive effort will have to be made by the population.

What is the problem with international taxation that these 130 countries have agreed to?

Multinationals have been playing countries off against each other for decades. They will structure their group in such a way that accounting profits are subject to the most favourable corporate tax regimes.  This practice was until recently considered a good entrepreneur's practice to reduce costs as much as possible in order to better reward shareholders for the risk they take.  However, over the years, this led to an ever-increasing shift of the tax burden from large companies to smaller ones, labour and consumption.

With the pandemic and the climate crisis, major expenditures should be made in everyone's interest and it is no longer socially accepted that larger players do not contribute proportionally in the countries where they accumulate their wealth.   In the earlier version of the second pillar, there was a danger that smaller countries with an open and export-oriented economy, such as Belgium, had to be concerned about. After all, groups then had an interest in concentrating costs (think of investments in research and development and highly qualified personnel) as much as possible, for accounting purposes, in larger countries where they had to declare more taxable profits. The incentives that smaller countries now offer to attract investment cannot match this financial logic without degenerating into prohibited state aid. By now working - in contrast to a previously proposed system - with outgoing payments as the basis of the corrections in the second pillar, this danger is now avoided.

By combining the effects of the second pillar (an effective tax burden of at least 15%) and the first pillar (profits above 10% are allocated to other countries up to a maximum of 20% to 30%), one could theoretically arrive at a level of taxation that would partially close the gap between digital and non-digital business models (5% to 6% more tax burden for digital activities). This reasoning is only valid insofar as payments are made from countries in which the users' data are collected. 

We have already indicated that one of the problems in the digital economy is precisely that the buyers of data are not necessarily located in the country of those users.  The effect of the second pillar may therefore prove rather limited if companies concentrate their payments in tax-friendly countries.  This also applies to non-digital business models.

The effect of the first pillar, as it is currently presented, seems to us to be too limited, because it works on the basis of booked profits and not on the basis of harvesting data and turnover.  We are therefore concerned that the introduction of the two pillars will not, in practice, lead to the hoped-for high increase in the tax burden on digital business models or significantly contribute to a better tax balance between digital and non-digital business models.

How does this agreement between 130 countries compare to the funding through a tax on digital activities of the European Union's post-pandemic and pre-Green Deal relaunch plan?

In 2018, the European Commission came up with proposals to tax well-defined digital activities of very large players uniformly in the European Union.   This led to the idea of a 'digitax' that you may have heard about. In 2019, Ireland and a few other Member States used their right of veto to block the European Directive that wanted to introduce the digital tax at the European level.  Other Member States (e.g. France) did this on their own but were soon threatened with trade sanctions by the Americans. From the American point of view, this whole regulation violates international agreements on the equal treatment of domestic and foreign companies competing in a country.  In Belgium, too, bills have been circulating since 2018 on the introduction of such a digital tax.  The agreement between the 130 countries demands that such digital taxes cease to exist.

Among the nine countries in the Inclusive Framework that do not share the agreement in principle, there are three Member States of the European Union: Ireland, Hungary and Estonia.  This means that these Member States, as long as they have not acceded to the agreement or transposed it into national legislation, can perfectly legally invoke the tax treaties with other countries to protect the companies established in their country from taxes that other countries would claim in implementation of the two pillars on which there is an agreement in principle between 130 other countries.  They can then claim unequal treatment or invite groups to organise payments from their countries.

The European Commission urgently needs its own funds for the Green Deal and to pay its own expenses for the recovery plan.   Following the agreement between the 130 countries on corporate taxes on digital and other activities and its acceptance by the G20, the European Commission has decided not to make a proposal to the member states on taxing digital activities within the European Union until October 2021. Presumably, the Commission first wants to wait and see whether or not the 130 countries will reach an agreement by October 2021 on the concrete implementation of the agreement in principle. The taxation of digital activities does not necessarily have to take the form of a minimum corporate tax.  From what has been leaked in the press about these plans, we can deduce that there would no longer be the same scope as in the 2018 proposals, i.e. no more high threshold (global group turnover of 750 million), nor the targeting of very specific digital business models.  The high threshold and the limited subject matter then meant a serious reduction in scope.  However, with the new proposals, the European Commission is targeting thousands of companies in Europe. In contrast, the agreement in principle between 130 countries still only affects a select club of large international groups.  Because of the large number of European companies that are or will be covered by the new regulation, the European Commission is reportedly hoping that the Americans will recognise that it is no longer only their successful companies that are targeted by the 'digital levy'.

However, Ireland, Hungary and Estonia, whether supported by baleful Visegrad Member States (Poland, Slovakia and Czech Republic) or not, or Member States wishing to prolong a favourable tax regime, can once again cripple or delay any European initiative on the digital levy through a veto.  The European Commission and the other Member States will then have a big stick in the form of scaling back the implementation of the recovery plan for those countries. 

Why, in our opinion, should Belgium now take its own initiatives until a fair and efficient mechanism is effectively in place at an international level?

The corona crisis has not only confronted us with a new high-pressure financing requirement, but has also shown an even larger proportion of consumers the way to e-commerce. With this 'candle burning on both sides', a tax on digital activities is a logical and urgent step to take to protect our companies and the budgets of our governments.

At present, the Belgian corporate tax system does not adequately and efficiently tax digital activities.  If the seller or intermediary has no physical fixed place of business in Belgium, the whole price or a significant part of the price escapes Belgian income tax.

Moreover, if that seller or intermediary pays less tax through tax optimisation or simply by applying lower tax percentages elsewhere, he can lower his prices to gain more market share in Belgium. There are also distortions in the area of social security due to the avoidance of paying Belgian wages for employee input that is the basis for turnover from digital activities in Belgium.  Fortunately, there is already an important correction in the area of VAT, which will level the playing field between competing companies in the Belgian e-commerce market as of 1 July 2021.

The companies that have a physical presence in Belgium will still have to compete on the Belgian market after 1 July from a very unequal starting position. This is an untenable situation, now that already two years ago (in 2019) at least 66% of Belgians made online purchases. This figure continues to rise year after year, and it is primarily young people (16-24 years) who are anchoring online purchases in their buying patterns, which will only reinforce the trend in the coming years. It is also to be expected that the closure of physical shops imposed by Corona will have generalised this phenomenon to other age groups.

Belgian legislative action is urgently needed if the concern is to ensure that, in budgetary terms, a fair share of this e-commerce contributes to the costs it generates (use of roads, consumption of electricity, closure of competitors) or if there is a real concern about major challenges such as maintaining social security and the urgent investment in climate objectives. A Belgian proposal must also create a more level playing field for Belgian traders. Once a supranational agreement becomes effective, whether at the level of the OECD or the European Union, a previously implemented Belgian system can be adapted to it. 

How can Belgium do this without having to amend treaties?

Relatively speaking, Belgium has the largest share of the population within the European Union that makes purchases via the Internet (Eurostat comparison). As regards goods, a large proportion of purchases escape income tax in Belgium because major players avoid a physical presence in Belgium. They are located in our European neighbouring countries: Zalando in Germany, bol.com in the Netherlands and amazon in France. Also Alibaba's distribution centre is in Maastricht (NL) and not in Belgium where the goods are flown in from China.

Double taxation treaties are now interpreted in such a way that a physical presence is required in order to subject sales in Belgium to the (relatively high) income tax. That is why so-called 'pure' digital players scrupulously avoid any physical presence. 

We advocate not waiting for a supranational agreement to enter into force to effectively tax digital activities in Belgium and have therefore developed a proposal ourselves. This proposal is fundamentally different from a bill already submitted (by cdH) that only subjects the turnover from certain digital activities to corporation tax.  Such a digital tax is seen by the USA as highly discriminatory and opens the door to trade sanctions. 

Our proposal is based on both EU law and globally accepted rules of customary international law on the interpretation of treaties to tax not on the basis of accounting profit but on the turnover from digital activities in the territory of a country.  The more costs or lost revenue that digital activity brings to Belgian society, or the more wealth that is accumulated through the harvesting of data, determines whether a larger or smaller part of that revenue can be taxed in Belgium at the corporate income tax rate and/or a special digital tax. The introduction of taxes on digital activities aims, in addition to alleviating budgetary needs, to perpetuate the social cohesion that the economic recovery will require, by ensuring a fairer and more efficient contribution from foreign and domestic companies that currently do not pay a minimum level of tax on their income from digital activities in Belgium.  Trade treaties, EU law and double taxation treaties all oblige Belgium to treat all companies equally.  There can therefore be no quick solution to the problem without imposing a minimum equal tax burden on both domestic and foreign companies for their digital activities towards Belgian users.

To this end, our extensive proposal targets business models that use digital activities with an increased risk of avoiding a minimum corporate tax burden in Belgium, or transferring the entire taxable base from digital activities in Belgium outside of Belgium by means of a payment to a foreign affiliate.  The own proposal thus follows the logic of the two pillars on which the 130 countries found an agreement in principle, but at the same time it differs from them because it works with more efficient and simpler criteria.  The criteria of physical presence and accounting profit will be completely obsolete in 2021 in order to achieve fair international taxation of wealth generated by digital activities.  The combination of a renewed corporate income tax and the introduction of an indirect tax on digital activities creates an efficient tax dam against BEPS (Tax Base Erosion and Profit Shifting) and against deloyal tax competition.

If a company already pays more than a minimum corporate tax on turnover generated by digital activities on the Belgian market, it does not pay extra in our proposal. Only those who do not currently contribute a minimum share are asked to make an urgent adjustment.   For the pandemic, we estimated the revenue from our proposal on direct taxes to be at least 1 to 2 billion euros. In addition, in the context of shifting fiscal and parafiscal burdens from labour to capital, and of the great cost of the green energy transition, it is requested that digital activities that fall under the minimum corporate tax also pay a levy on their turnover in Belgium according to the electricity consumed by the user and the possibility of harvesting his/her data. For the pandemic, we estimated the revenue from our indirect tax proposal to be between EUR 4 and 8 billion, depending on the criteria chosen. This whole new Belgian mechanism can be introduced perfectly alongside the European initiatives and the agreement in principle between 130 countries.  It would allow for an intelligent way of distributing the tax burden more fairly in Belgium, not only between companies but also between companies and families.

The great advantage of our proposal lies in the fact that it does not require an international agreement, treaty changes or adaptation of directives.  It can therefore be implemented immediately. This means an enormous gain in time and considerable additional revenue in the short term. The proposal also avoids discussions about different accounting rules in countries or different treatment between domestic and foreign companies, which could lead to a treaty violation, which may or may not be permitted by the treaty partner.  The mechanism we provide does not require assessing the application of tax rules of other countries, which significantly reduces the risk of trade sanctions or diplomatic complications. In addition, the path we have proposed removes the need for a complex dispute resolution mechanism. The proposal does not need complex accounting rules for the very large digital players, which is still a pitfall in international proposals.  The criteria used by us are clear and equal for all players who wish to harvest digital data in Belgium and sell it here via the Internet. The proposed mechanism leads to an immediate and effective minimum tax of between 4.9% and 8.05% on the turnover from digital activities, depending on the digital activity in Belgium and depending on the additional costs that our governments have to bear or the revenue that they lose through the mere fact that existing companies are squeezed out or through tax optimisations. 

The agreement in principle between the 130 countries is important because it recognises the right problems. We hope that by October 2021 there will be a growing awareness that the solutions adopted can be improved. From the European Commission's work and positions on 'the digital levy' leaked into the press, we can at least deduce that equal treatment is an essential starting point for the Commission.  Precisely because our proposal is also based on this same fundamental principle, we were looking forward with interest to a concrete European proposal that would be published in July.  The fact that the European Commission is now postponing the presentation of the 'digital levy' until October to see how the agreement in principle works out in the meantime, while at the same time saying that it might no longer introduce it, is possibly a negotiating technique for the discussions between the 130 countries. The European Union still needs its own resources to repay the loans it has taken out on the financial markets. 

The conclusion is that, as of today, there is no certainty any more about a European initiative and the agreement in principle between the 130 countries will most likely not be reached in a concrete form before 2022 or 2023, after which it still has to be applied... We believe that Belgium cannot afford, in budgetary terms, to wait until 2024 or longer for any international proposal whatsoever.  It is time to act here and now and, in a spirit of loyalty, to adapt the new Belgian regimes to the changed international context.  In other words, one cannot satisfy one's budgetary hunger with ambitions that are not concretised and, above all, not fixed ("voted"). It is better for the federal policymakers to retreat into the Belgian kitchen for a while. Otherwise, there is a risk that little will be served this year and the years to come.