Go back to menu

New UK taxes on offshore IP structures - UK Budget 2017

The impact on traditional and digital businesses

24 November 2017

The Government's Budget was announced by the Chancellor on 22 November 2017. It includes a series of proposals attacking offshore IP structures. The main targets are foreign digital businesses, but all businesses that employ offshore IP structures will be affected. The Government's long term objective is a new multilateral tax applied to social networks, marketplaces, and other businesses that generate value from user participation. However, as a short term interim measure, the UK will be introducing a unilateral extra-territorial withholding tax on offshore royalty payments to tax havens, when the payments are ultimately funded by UK sales. This briefing summarises the new proposals, and looks at the steps business can be taking to prepare for the new taxes.

What is the political objective? 

The last few years have seen increasing controversy over the way in which many large US-headquartered digital companies are taxed on their worldwide non-US profits. 

Typically they pay very little tax outside the US. 

  • All sales outside the US are booked by entities in countries such as Ireland, Luxembourg or The Netherlands. 
  • These sales entities contract directly with consumers/sellers/ advertisers (depending on the underlying business model). 
  • The arrangements are carefully structured so the sales entities have no taxable presence ("permanent establishment") in the countries such as the UK where the consumers who buy the products reside. 
  • The sales entities' considerable revenues are then offset by large royalty payments to IP holding companies in low tax jurisdictions 
  • The US would tax the resulting profits if they were repatriated to the US; however they are kept offshore and remain untaxed. 
  • In principle this is merely tax deferral, but in practice the deferral is often permanent (although that may be impacted by the current proposed US tax reforms). (This deferral may be disguised in accounts by way of a large deferred tax provision being made against the often remote possibility the profits are returned to the US.) 

Many have criticised these kinds of arrangements as unacceptable tax avoidance. Whether that is the case is debatable. There is a good argument that the profits should be taxed in the US, and not Europe, given they arise from IP, technology and brands that were developed almost entirely in the US. The fact the US in fact chooses not to tax these profits should not change that point of principle. 

The additional complication is that in many cases the profitability of digital businesses is limited or even zero – Uber, for example, lost c$700m last quarter. Expecting such businesses to pay material tax on their profits is therefore somewhat misconceived. 

Observers might also make the point that for a century or so, most European countries have had tax systems that levied tax based on the concept that the headquarters ("brains") of a company is a significant source of its profit - so that a European company selling to the wider world would be taxed in Europe. It is not entirely logical to now reject this orthodoxy just because the brains of companies are often not in Europe, and in the digital economy it is relatively easy to move the brains to low tax locations. 

However these arguments have little political salience in the current environment. Many people feel intuitively that a business that generates value from users in Europe should be paying significant levels of tax in Europe. That feeling is perhaps heightened by Europe's relative failure to generate world-beating digital businesses. For whatever reason, there is now perceived to be a public loss of faith in the corporate tax system, and policymakers are looking to new digital taxes as one potential solution.

What previous attempts have been made to tax digital businesses? 

There have been several attempts to attack the structuring summarised above. This has, however, been complicated by the fact that the UK and other countries are trying to solve a problem that is (in our view) fundamentally created by the US tax system. 

One approach is to attack the arrangements that prevent a taxable presence arising in the UK and other "source" countries. The first example of this was the UK diverted profits tax (see our briefing). This was followed by BEPS Action 7 (see our briefing). However, these measures are generally perceived to have been of limited effect on digital businesses, and have raised little in additional revenues. 

The problem is that the conventional approach to tax adopted by OECD countries only permits a country to tax profits attributable to local personnel and assets – and the UK personnel and assets of many digital businesses are relatively limited drivers of UK profitability. 

2017 has therefore seen policymakers move away from traditional profits taxes and look at alternatives. 

Perhaps out of frustration with the status quo, the Commission has pursued innovative theories of State aid to try to retrospectively undo some of the tax rulings that underpin the kinds of structure summarised at the start of this note. However, the legal basis for this approach is highly contentious. 

Long term multilateral alternatives 

Several alternative ways of taxing digital businesses have been proposed. 

The most radical proposal is to entirely replace the current system of taxing corporate profits. Instead of each entity in a corporate group being taxed separately, the group as a whole would be taxed, with its profits allocated between the different countries in which it operates on the basis of a formula that takes into account sales, number of employees and assets. 

This approach - "unitary taxation" or "formulary apportionment" - is backed by many NGOs, some academics and the European Commission, but has limited support from national governments. That seems to us an insuperable bar to adoption, given its effectiveness depends on near-universal implementation. 

The UK continues to dismiss unitary taxation as a general matter. However, the HM Treasury paper does take a significant step in that direction. It proposes that, where a business generates value from user participation, then governments should have a right to tax a portion of the profits that reflects the value generated by their local user base. The rationale is that "user participation" is currently given no weight in the allocation of taxable profits between entities and countries, and that any solution should therefore specifically address that concept. 

So, for example, the proposal would give:

  • France the ability to tax a portion of Facebook's profits based upon the number of French users; 
  • Germany the ability to tax a portion of eBay's profits based upon the number of German buyers and sellers; and 
  • the UK the ability to tax a portion of Airbnb's profits based upon the number of UK hosts and guests. 

This proposal has the considerable merit of taxing the desired target – international digital businesses – based upon the fundamentals of what they do, as opposed to whether they happen to provide a solution online that could equally be provided offline. That therefore avoids the economic distortions, difficult definitional questions and potential avoidance that would follow an attempt to define and tax "digital businesses" as such. 

Furthermore, one important difference between this proposal and unitary taxation is that it remains viable even if a significant number of countries fail to adopt it. 

We expect it would be a new tax (and not an extension of existing taxes on corporate profits) and would therefore fall outside most countries' double taxation treaties. 

There are, of course, obvious challenges. 

− First, the need to define, and reach an international consensus, on what is meant by a business that generates value from user participation. 

− Second, the practical challenges of how such a tax would be assessed and collected. We would expect the most well known internet businesses to comply voluntarily; others may not. 

− Third, the tax would not apply to digital businesses that sell their own digital or physical products over the internet (as they do not "generate value from user participation". This may be felt by some to be inadequate. 

− Fourth, the US is most unlikely to participate in any such new tax (for a variety of political and policy reasons), and may seek to frustrate it – or even impose retaliatory measures. 

− Fifth, history suggests that radical tax policy often leads to unanticipated consequences. The effects of the UK "window tax" introduced in 1696 can still be seen in blocked up windows. 

The new tax could create double taxation, given the same profits would be taxed by conventional taxes in a company's home jurisdiction and by the new tax in user jurisdictions. This problem may, however, be limited in practice, given that the principal targets of the tax will be US-headquartered businesses, and a combination of US tax law and careful structuring means that they may never pay material US tax on their non-US profits. 

In making this proposal, the UK could be accused of trying to have its cake and eat it. The UK applies a form of unitary taxation to digital business where the UK has struggled to compete, but maintains its opposition to unitary taxation in those industries where UK businesses have a significant presence.

Interim multilateral solutions 

Until agreement can be reached on a multilateral solution, the European Commission and the OECD have each proposed several possible alternative interim measures, with the Commission promising implementation across the EU if multilateral agreement cannot be reached. 

The problem is that none of the options presented so far look very attractive: 

  • An equalisation tax on the turnover of digital businesses. The idea seems to be to take the profits that are currently un-taxed (or insufficiently taxed) and apply a tax to them in countries where a business has a "significant economic presence". This is problematic in principle – if the US chooses to not tax profits then why should that automatically make them taxable in Europe? It also begs the question of what it means to have "significant economic presence". 
  • A withholding tax on digital transactions. It seems distortionary to tax gross revenues and not profits, and the incidence may well fall on the end user. Furthermore whilst one might expect businesses to withhold tax, it seems unrealistic to expect consumers to do so. 
  • A levy on revenues generated locally from the provision of digital services or advertising activity. This would again seem to be a distortionary turnover tax, with the benefit of not relying on a withholding mechanism, but the additional challenge of the EU law prohibition on turnover taxes. 

The UK is now proposing that the OECD adopt a variant on the third of these options, but aligning with the specific "user participation" concept. 

That seems to us a slightly strange approach to take – once the hard work of agreeing the "user participation" concept is complete, why not jump straight to the long term solution? 

The UK's own interim unilateral solution 

Given that all the above proposals will require multilateral agreement and implementation, the UK Government is proposing an interim unilateral solution, targeted specifically at the kind of structuring summarised at the start of this briefing. 

It would do this by applying a new UK withholding tax to royalties: 

− paid by a company, anywhere in the world 

− to a no or low-tax jurisdiction with which the UK doesn't have a tax treaty (which typically means the kind of jurisdiction that is often referred to as a tax haven) 

− in connection with sales or products and services to UK customers. 

This is a much simpler approach than the kind of digital taxes discussed above. In a straightforward case, it is easy to apply. 

If we assume the rate of the withholding is the usual UK rate of 20% then the tax advantage of the example structure summarised at the start of this briefing is more than reversed.

We can see how this simplicity is attractive. However it carries with it a number of challenges: 

High effective rates, leading to potentially low yields. 

In the example above, the tax equates to an effective rate of almost 20% on the gross revenues of the company – but the effective tax rate on profits would be much higher. The most profitable digital businesses have a profit margin of around 40%: the proposed tax could therefore equate to an effective 50% tax on their profits. For a more typical digital business, the effective rate could easily exceed 100%. Doing business in the UK would be a loss-making endeavour. 

The obvious response to this would be for such companies to abandon their offshore structures and have royalty payments made directly to the US, or via a jurisdiction with which the UK has an appropriate tax treaty. There are a number of ways this could be done, and the overall effective tax rate would likely then drop to a maximum of 35% (under current US tax law) - and possibly a much lower figure. Note that in some such scenarios, not a penny of additional tax would be collected in Europe – the only winner would be the US Treasury. 

Another response would be for the UK to set the rate of withholding tax at a level that is sufficiently modest to not prompt wholesale re-engineering of international structures. That is, however, difficult, given the wide disparity in profitability between different digital businesses. 

The Budget "Red Book" shows the tax raising only £200m – a small fraction of the UK revenues of digital businesses. This suggests that the tax is indeed intended to change behaviour rather than raise revenue. In a sense this is a tax that is designed to be avoided. 

Impact on other businesses 

The paper suggests that the withholding tax will not be limited to "user participation" businesses, but will apply to all businesses that hold IP in tax havens and pay royalties linked to UK sales. That will include many businesses that are not primarily digital but which rely on IP. Some obvious examples are pharmaceutical and music groups; but in the modern world most large businesses of all kinds have a significant IP component, and this is often held offshore. Even consumer goods, retail and financial services businesses often build structures around their very valuable IP. 


The example above has a very clear link between UK sales and offshore royalty payments, but real life cases may not be so clear-cut. Some businesses might also seek to avoid the tax by recasting royalty payments

into another form, e.g. interest, derivative payments or profit share payments. We would expect, as a result, complicated legislation and anti-avoidance provisions. 


It is not obvious how HMRC would be aware of offshore royalty payments, or how an extra-territorial withholding tax could be collected. This is unlikely to be a problem in practice for the very large digital businesses, who we expect would in the main be self-policing. However, smaller players may in practice be difficult to tax – and Brexit may make cross-border enforcement more difficult for HMRC than it has been in the past. 

EU law 

The practical effect of the proposal is to discourage the movement of IP to low tax jurisdictions. It is possible that this effect contravenes the EU law freedom of movement of capital, as this applies to movements of IP, and is not limited to transactions within the EU. Such considerations may be moot after the UK leaves the EU in (most likely) March 2019 – but if there is to be a transitional period then EU law challenges are a distinct possibility. 

When will this happen? 

The Government will release a consultation paper on the unilateral withholding tax shortly, with implementation from April 2019. 

That unilateral implementation might never happen if agreement can be reached on interim multilateral solutions (although that will be extremely challenging). These are likely to be the subject of extensive international discussions in the first quarter of 2018 with implementation following as early as 2019 (if agreement can be reached). 

The timescale for long term multilateral solutions is entirely unclear at this point. 

What should business be doing now? 

Any business which makes royalty payments to offshore IP-holding companies will be potentially impacted by the proposed new withholding tax. It would be prudent at this point to identify the potential exposure and begin high level thinking around alternative structures. 

Restructuring IP holding/royalty arrangements will require a complex balance between intellectual property law, US tax and European tax considerations. Given the ongoing US tax reform process it is likely premature to progress structuring past the very conceptual. However, businesses may wish to move quickly once the outcome of US tax reform becomes clear. 

Further Reading

Corporate tax and the digital economy: position paper

If you would like further details on any aspect of this article, or how it applies to your business, please speak to your usual Clifford Chance contact or any of those listed below.