Will the OECD’s global tax deal raise more from tech companies in the UK?

27th September 2021

Analysis shows that tech giants would pay less in taxation in market jurisdictions under G7 global tax deal than under current digital services taxes.

Proposals put forward by the G24 group of nations ahead of next month’s crucial meeting of the OECD would lead to tech companies paying more.

Making tech companies pay their fair share?

In June 2021 the G7 reached an agreement on a new system of taxation for the world’s largest companies.

The deal unlocked ongoing negotiations at the OECD and was billed as a way of confronting the “tax challenges that arise from the global digital economy” and a way of making sure multinationals pay “their fair share in the countries they do business”.¹ Final sign off on a deal at the OECD is expected in October.

The agreement was the product of years of negotiations at the OECD following public outcry at the low levels of taxation being paid by large tech companies – some of which had seen their effective tax rates fall to single digits.

Although tech companies have insisted that the low tax rates they have achieved have been a result of government incentives, the reality is that the industry was a heavy user of tax avoidance schemes.

Talks aimed at addressing the problem had been taking place at the OECD for many years. In the interim, many countries around the world grew increasingly frustrated with a lack of progress and started to implement their own unilateral measures to tax digital companies and “make them pay their fair share”.

These so called “digital services taxes” (DSTs) were a controversial measure that sparked trade disputes between the United States and European countries that adopted them.

One particular aspect that aggravated tech companies was that DSTs operate as a tax on the turnover in jurisdictions where services were sold. This meant that they did not come under tax treaties and therefore could not be credited against taxes paid in a company’s home jurisdiction in the say way that corporate taxes can be offset against liabilities at home. They were a tax that came on top of any corporate tax liability.

From the point of view of tech companies, a new global system of allocating taxable profit to jurisdictions to replace a patchwork of DSTs would be easier for tech companies to administer, allow them to manage their tax liabilities on a global level, and have the added benefit of wiping the slate clean following years of tax avoidance.

For this reason, tech companies have long argued for tax reform and were supportive of the G7 deal when it was announced. A spokesperson for Google told Reuters: “We strongly support the work being done to update international tax rules. We hope countries continue to work together to ensure a balanced and durable agreement will be finalised soon.”²

Nick Clegg, Global Vice President of Public Affairs for Facebook Tweeted: “Facebook has long called for reform of the global tax rules and we welcome the important progress made at the G7. Today’s agreement is a significant first step towards certainty for businesses and strengthening public confidence in the global tax system. We want the international tax reform process to succeed and recognize this could mean Facebook paying more tax, and in different places.”³

But will companies like Facebook end up paying more tax in market jurisdictions like the UK?

In this paper, we demonstrate that for companies subject to the DST rate of 2%, the DST rate in the UK, it is a near certainty that they will end up paying less in tax in “market jurisdictions” such as the UK under the G7 proposals than under existing digital services taxes. If the UK supported the position of the G24 group of nations, more profit would be allocated to the UK, with the result that profit margin required before a company pays out more tax under the OECD formula than the current DST being closer to the profit margins observed at large tech companies.

We come to this conclusion based on a mathematical analysis of the formulas used to calculate the DST and the new tax base under the OECD proposals.

The effective tax cut that the G7 deal represents is even larger for countries with higher DSTs like France.

The G7 deal

The agreement reached at the G7 was based on a blueprint put together by the OECD.

The new proposals come in two parts, Pillar 1 and Pillar 2. Pillar 1 is a mechanism to re-distribute profits made by some of the world’s largest multinational companies to countries where they make revenues (such as the UK). Pillar 2 is a mechanism to stop companies from moving their profits into low tax jurisdictions. In this paper we focus on the Pillar 1 proposals. Under Pillar 2 all of the benefit goes to the country where the multinational is headquartered, and as such it is not relevant to the distribution of profit to market jurisdictions.

The OECD blueprint outlines a three step process to find the “allocation amount” under Pillar 1, which is the amount of profit that will be available to market jurisdictions to tax.

The G7 deal largely adopted the blueprint, and agreed the various percentages required to determine the allocation amount, which had been left as an open question.

Following the G7 agreement, we can now see what the impact will be on the tax base of large corporations using the blueprint process and the allocation percentages set by the G7.⁴

Step 1 – Allowance for “routine” profit

The first step in the OECD process is to isolate what the OECD call “routine profits”. This applies a simple percentage to the profit before tax of a multinational as an allowance that will excluded from any reallocation. The G7 agreement set this as 10%.

Step 2 – Global Reallocation

The next step in the process is to create a pool of profits that will be available for reallocation to market jurisdictions. This is a simple percentage of the remainder of a company’s profit before tax after step one. The G7 set this reallocation percentage at 20%.

Step 3 – Redistribution

The reallocation pool arrived at after step two is then carved up amongst market jurisdictions who can then apply tax the profits allocated to them.

The G7 agreement did not give any details on how this would work, however, the OECD blueprint states that the key will be based on the revenues a Multinational Enterprise(MNE) makes in any in-scope country.

For the purposes of this paper, we have assumed that the allocation is based on a simple proportion of total revenue made in any country.

The three steps can be expressed as a formula as follows

Local tax base = Global revenues x (profit margin – 10%) x 20% x local market share

This local tax base can then be multiplied by the local corporation tax rate to find the amount of additional taxation raised by the pillar 1 process.

Digital Services Taxes

As part of the G7 agreement reached on the new global tax deal, countries should dismantle domestic digital services taxes. The rational for this is that the DST was a stop gap to ensure that some tax was paid in market jurisdictions until a global agreement of the kind proposed by the OECD was agreed.

Digital services taxes are crude and simple taxes that place a flat rate of taxation of the revenues of a company derived from the provision of certain digital services in the country.

In the UK, the DST is a tax of 2% on the revenues derived from search engines, social networks and online marketplaces with a tax free allowance of £25,000,000

The UK DST can also therefore be expressed as follows:

DST revenues = (Global revenues x local market share) -£25m x 2%

Comparing DST revenues with Pillar 1 revenues

For companies subject to the DST, i.e. those that derive their income from in-scope activities, we can easily compare the amount of tax raised by the DST against the new pillar 1 formula. We have already done this exercise using the real corporate accounts of several global tech giants and our estimates of revenues in the UK here.

However, we can also determine under what conditions the DST will equal tax revenues raised under pillar 1 because by putting the two equations together we get:

(Global revenues x local market share) -£25m x 2% = Global revenues x (profit margin – 10%) x 20% x local market share x local tax rate

We can see here that for any given company, their global revenues and local market share will be the same on each side of the equation, meaning that the only two variables that will change whether or not the DST raises more than Pillar 1 for any company subject to the DST will be determined by the profit margin of the company and the local tax rate.

This equation can therefore be rearranged to isolate profit margin, allowing us to answer the question, what profit margin would a company need to achieve in order to end up paying more under the pillar 1 than the DST at a given local corporate tax rate?

Profit margin = ( (Global revenues x local market share) -£25m) x 2%/ (Global revenues x 20% x local market share x local tax rate) + 0.1

If we take a company that has global revenues of £10bn, and a market share in the UK of 10% of global sales, and a UK corporate tax rate of 19% we get the following:

Profit margin = ( (£10bn x 0.1) -£25m) x 0.02 / (£10bn x 0.2 x 0.1 x 0.19) +0.1

Which gives us

Profit margin = 62%

So, under the current UK corporation tax rate of 19%, a corporation would have to achieve a pre-tax margin of 62% before it started paying more under the pillar one agreement than under the digital services tax.

The UK corporation tax is currently scheduled to increase to 25%, so we can re-run the exercise at that tax rate with the same notional £10bn company.

Profit margin = ¹£10bn x 0.1) -£25m x 0.02 / (£10bn x 0.2 x 0.1 x 0.25) +0.1

This gives us

Profit margin = 49%

The equation is highly sensitive to increases in the DST rate. If we look at the same company, and apply the French DST rate of 3% and the French Corporate tax rate of 28%, then the profit margin required is 68%.

Real profit margins of tech giants

Of course in theory, a company could end up paying more under the pillar one process than under the digital services tax if it achieved very high profit levels.

However, a quick glance at the real historical profit margins seen by tech giants demonstrates that the kinds of profit levels required before a company ends up paying more under pillar one are rare. As such, it is a near mathematical certainty that companies will end up benefiting from a deal that removes digital services taxes and replaces them with a Pillar 1 allocation under the deal announced at the G7. This is a highly significant finding, given that the purpose of the G7 deal was to come to a more equitable taxation of the digital economy.

The following chart gives us the historic pre-tax margin of Google, which has never risen above 37%.

The following chart is the pre-tax profit margin of Facebook. Facebook’s profit margins have been as high as 53% in the past, but more recently have hovered around the 40% mark.

Source: macrotrends.net

The G24 proposals

The fact that the G7 proposals involve very little distribution of profits away from the home jurisdiction of multinational companies (most of which are headquartered in G7 countries) has not gone unnoticed.

The G-24 group of developing nations has put forward alternative proposals to those adopted by the G7.⁵

The most signifiant change to pillar 1 is to the allocation percentage, the percentage of a multinational’s profits (above the 10% hurdle) that would be available for reallocation to market jurisdictions. Under the G24 proposals this should be at least 30%.

If we rerun our formula above, substituting an allocation percentage of 20% to 30%, we find that assuming a 2% DST and a 19% corporate tax rate, a company would only have to have a profit margin of 44% before Pillar 1 started to raise more revenues than the DST. With a corporate tax rate of 25% that falls to 36%.

It is therefore in the best interests of the UK to support the proposals put forward by the G24 group and seek a higher allocation percentage.

This report is also available as a PDF here.

This article was amended on 27th September to correct a mistake in the calculation of the impact of the G24 proposals.