There is widespread international consensus that the current system for taxing multinational companies is broken beyond repair. The OECD is currently negotiating a new system, with the organisation aiming to come to an agreement early in 2020. The current OECD proposals move away from the traditional approach of taxing (or attempting to tax) multinationals, and introduce an element of unitary taxation. The position of the UK government in the negotiations over the OECD proposal has become an election issue, with the Labour party announcing that it wants to go much further and introduce a fully unitary system, taking unilateral action to deliver it if necessary.
What is unitary taxation and how do the OECD’s proposals work? In this article we set out some of the issues confronting politicians as they seek a solution to the problem of tax dodging by multinational enterprises and why the OECD’s only partial adoption of the unitary principal is unlikely to solve the issue. The article then goes onto deal with some of the criticisms of the unitary approach, arguing that it is possible for the UK to take action on a unilateral basis which would move us towards a unitary system, even if international agreement on a fully comprehensive global reform was not reached.
A broken system
In 2016 Google made almost exactly 50% of its pre-tax profits in the United States and 50% in the rest of the world ($12bn and 12.1bn).1 On that profit, it paid $3.8bn in US federal and state taxes, and a further $966m was divided between every other government in the world where Google operates.
The reason for this extreme inequality is that in 2016 Google Ireland Holdings Unlimited, a company with an address in Dublin but tax resident in Bermuda, declared a profit of $8.9bn which was subject to a 0% corporation tax rate.2 As stated in the Google 10-K form, this was most of the profit that the company made outside of the United States (although the 10-K uses the term “substantially all”). Google Ireland Holdings Unlimited appears to have no employees, and all of its income derives from the ownership of intellectual property rights.
For 8 years governments at the OECD have been trying to agree how to reshape the rules around international tax. A number of reforms have already been put in place under the BEPS process, but as the OECD itself accepts, these have failed to grasp the kinds of structures put in place by Google and others.
This is confirmed by our own research, which found that the effective tax rates in the UK of five of the largest tech corporations in the UK had barely changed following the last round of BEPS reforms.3
The choice facing governments
Governments are now facing a choice. Do they defend the individual entity principle, often called the arm’s-length principle, the current foundation of the international tax rules? This states that the hundreds of thousands of individual companies that make up multinational corporations should all be taxed as independent businesses. This approach has facilitated the current tax structures of multinational companies, which rely on trading between group companies to move profit from one jurisdiction (such as the UK) to another (such as Bermuda).
The alternative approach is unitary taxation, which treats multinational corporations as one single enterprise, dividing the right to tax their total global profits between the nations where they operate based on a formula. This formula could be based on sales and employees for example, ensuring that companies only face tax liabilities in countries where they have substantial operations. The system already exists in the US, where it is used to divide taxable profit between different state governments which levy their own corporation tax rates. The Labour party have announced they will implement a unitary system as part of their platform in the 2019 election.
As a matter of pure principle, the unitary principle must surely be the correct approach. As was set out long ago by Lord Carnwath LJ, tax statutes ‘draw their life-blood from real world transactions with real world economic effects’. In the real world, multinational groups are single businesses rather than a set of loosely affiliated, separate entities.
However, there are many who still cling to the individual entity myth. Why? We can only speculate. However, it cannot go unnoticed that the individual entity approach generates mountains of paperwork for each multinational corporation, as each individual subsidiary must file its own accounts and justify the transactions it makes with companies within the group. This creates large amounts of work for accountants and tax advisors, all of whom are paid hefty fees for their work. These firms of accountants are often the same people acting as the voice of business when it comes to international tax reform.
The OECD’s hybrid approach
The current proposals from the OECD manage to avoid taking a decisive position on the choice between the arm’s-length system and the unitary system. Instead of throwing out the individual entity approach, the OECD have proposed a hybrid system. The arm’s-length principal stays, but a new right to tax a “residual” profit is added to the international tax system. The residual is apportioned to countries on the basis of where a company makes sales. What comprises the “residual” is not made clear and is the subject of ongoing negotiations.
Whilst the willingness of the OECD to consider moving away from the current tangled mess created by the arm’s-length principle is to be commended, putting a unitary system on top of it will only add difficulties rather than solve them.
The first issue is who will be subject to the new tax. Under the OECD proposals, the new taxing right will not be targeted at all multinational businesses, only “consumer facing” businesses will be in scope.
A test which only included purely consumer facing businesses would exclude companies like Google and Facebook, who derive the majority of their revenues from advertising. Overwhelmingly the purchasers of advertising are other businesses and not consumers.
The OECD accepts this and talks about consumer facing elements, which would encompass online advertising businesses. To stretch the definition of consumer facing in this way could end up creating significant grey areas which could be open to exploitation. The private equity industry has demonstrated very well that the profits of almost any industry can be easily converted to finance income for example.
It is also the case that most big tech companies are no longer purely tech companies. Their sheer size is seeing them expand into driverless cars, banking, and other industries. Is the proposal to separate out the ‘digital consumer facing’ parts of these companies? Where and how will that line be drawn?
It would be better to broaden the unitary principle to a wider set of companies rather than simply seeking to place companies or parts of companies into an artificial consumer facing category.
The second question is what will be taxed?
In order to defend their maintenance of the arm’s-length approach within their proposed hybrid system, the OECD states that arm’s-length remains a good foundation for the division of taxing rights with regard to most transactions. The new unitary approach will only apply to the “residual”.
The evidence we have seen demonstrates that currently businesses are able to eliminate not just a residual but most if not all of their profit in market jurisdictions using the arm’s-length approach.
As demonstrated by the Google example given above, despite substantial marketing and research operations in the UK, the outcome of the arm’s-length approach is, by Google’s own admission, substantially all of their non-US profit being moved to tax havens. If the purpose of the new system is to plug the gaps in the current system, is it really credible that substantially all of the profits a company makes in Europe could be considered a ‘residual’?
TaxWatch has also studied the video games industry in the UK. Our work discovered that the makers of Grand Theft Auto have not declared any profit in the UK over the last ten years, despite producing the most popular entertainment product in history.4
This perverse outcome was achieved because Take-Two Interactive, the US headquartered multinational which publishes Grand Theft Auto, reimburses the UK based developer of the game at a rate which means that it does not make a taxable profit in the UK.
Under the arm’s-length principle the contract between Take-Two and its own UK based developer, Rockstar North, should be the same as if Take-Two had commissioned an independent developer to do the work for them. It does not seem credible that an independent contractor would accept such a bad deal for creating such a cash cow for its employer. As we found, the real pay agreements entered into between Take-Two and senior staff at Rockstar confirm this. Under the agreements, senior staff at Rockstar Games received a bonus based on the profits made by the games they developed (on the basis of the profits made by the Rockstar group as a whole), an arrangement which was not available to the company they worked for. Many other companies operating in the UK have a similar structure, where employees are compensated on the basis of the performance of the entire multinational enterprise, but the company they work for is not.
Despite this, as far as we are aware, Take-Two’s business model has not been challenged under any existing rules.
Further evidence of how the arm’s-length principle is not working is provided by HMRC. In February 2019 the UK government announced a new Diverted Profits Tax Compliance Facility.5
The background note published alongside the facility noted:
“HMRC has found that some Multinational Enterprises (MNEs) have adopted cross border pricing arrangements which are based on an incorrect fact pattern and/or are not consistent with the Organisation for Economic Co-operation and Development (OECD) Transfer Pricing Guidelines (TPG)”
(TPGs set out how the arm’s length principal should be applied)
Although HMRC will not say how many companies they are targeting with this scheme, the fact that the UK government created a new compliance facility to deal with such MNEs suggests that the problem was significant, and not isolated to a few companies. It also suggests that even in cases where companies can be challenged under the existing rules, tax authorities are struggling to deal with the issue.
Again, it is not clear how the introduction of a unitary element will solve the inherent problems with the arm’s-length principle, or why it would not be better to subject the entire tax base of a multinational corporation to the unitary approach.
Can unitary taxation be introduced without international agreement?
Those that oppose the introduction of a unitary system claim that the solution is an overly simplistic answer to a complex problem.
In particular, they argue that such an approach conflicts with existing tax treaties, all of which would need to be renegotiated. To implement a unitary system would require global agreement, something which would be impossible to achieve.
Of course, the ideal outcome would be for governments to agree on a comprehensive solution. But that does not mean that nothing can be done until that happens. As has been set out in the recent paper by Professor Sol Picciotto (Taxing multinationals: a new approach), many tax treaties already in force already permit the use of a profit split methodology to determine taxable profit attributable to permanent establishments. This allows countries to attribute the global profits of a multinational to offices located in their jurisdiction on the basis of a formula. The OECD accepts that this could provide a ready made legal basis for countries adopting an apportionment method. In addition to this, there are new and long standing anti-avoidance initiatives which move towards a unitary approach. Full inclusion CFC rules are one example, as are the much newer US GILTI provisions.
It could also be argued that in order to stem the clearly abusive schemes that are commonplace today, more drastic action is required. It is possible for Parliament to legislate to unilaterally disapply the provisions of tax treaties. This last happened in the UK in 2008, when the government legislated to unilaterally override all of their tax treaties to close down a disguised remuneration scheme (FA 2008 ss 58 and 59, which amended ICTA 1988).
The action by the UK government was subsequently upheld by the European Court of Human Rights, which noted that double tax treaties should do no more than seek to relieve double taxation, and should not be permitted to become an instrument of avoidance.6
We have proposed that the UK government take this approach with regard to the taxation of royalty payments, a measure which would attack the current profit-shifting models of the tech companies and bring in a potential revenue of £8bn a year.
One of the drivers of the OECD talks has been the willingness of governments around the world to consider unilateral action such as new digital services taxes. It is thought by many that these new taxes are problematic under the current international tax rules. This is one reason why the UK only announced a DST as a backstop if the OECD talks fail.
In this context it makes sense for the UK to continue to make the argument that if the OECD cannot find agreement on how to end tax dodging, it will take action itself. This may not mean a global comprehensive system of unitary taxation, however, there are many tried and tested anti-avoidance mechanisms which could be brought introduced by any government under the rules that exist today. These would move us towards a unitary approach even if none was agreed at the OECD and would be an effective means of closing down avenues to profit shifting by multinational companies.
1 Google’s 10-K form states pre-tax earnings in the US and International divisions, the 2016 results are used as 2017 was an anomalous year due to one-off charges associated with the US tax reforms.
2 The annual report of Google Ireland Holdings Unlimited Company for 2017 is available from the Irish Corporate Records Office
6 Huitson vs United Kingdom