In July, more than 130 of the world’s major economies agreed a landmark deal to change the international tax system to address the tax challenges arising from the digitalisation of the economy. The countries agreed a global minimum tax rate of at least 15% and the reallocation of some taxing rights to countries where multinationals do business.
Under the current system, companies operating in another country are taxed in that other country to the extent that their profits are attributable to a permanent establishment there (i.e. a branch or dependent agent). This concept dates back to the 1920s when companies could not do significant business overseas without having a fixed base and personnel there. The internet has changed this, but the tax system has not kept pace.
Tech companies have been able to generate substantial revenues, particularly from the sale of advertising which targets customers in the countries where they operate (so called ‘market jurisdictions’), but – at least in the view of local tax authorities - without paying very much (if any) direct tax in the market jurisdictions because they do not need a fixed base there to generate revenues. Governments in market jurisdictions have been understandably keen to try to get a share of the tax revenues of these highly profitable companies. Public opinion in these countries, particularly in Europe, has also been very supportive of attempts to make these (often US owned) multinationals pay their ‘fair share’ of taxes in the countries where they operate. This has resulted in a proliferation of unilateral digital sales taxes. These have been imposed on the revenues of social media and other digital companies by countries including the UK and France.
Over the last few decades we have also seen offshore financial centres competing to attract multinationals with low corporate tax rates. Previous high rates of US tax for overseas profits repatriated to the US, in particular, encouraged complex offshore ownership structures for the overseas activities of US owned multinationals, resulting in low levels of tax and (sometimes) profits escaping tax altogether. The countries where the multinationals are based (and the US in particular) consider they are losing out on tax revenues as a result of this ‘race to the bottom’ of corporate tax rates. Seeing profits derived from activities in their jurisdiction taxed at very low rates has also encouraged market jurisdictions, typically in Europe, in their demands for a share of the tax revenues.
Agreement on a solution has been difficult to achieve because of the differing vested interests of the countries concerned. The US thinks it should get the tax revenues from US owned multinationals, market jurisdictions in Europe (where a large number of technology users are based) think they should have a share of the revenues, and lower tax jurisdictions such as Ireland want to maintain their low rates of corporate tax to keep the jobs and economic activity they have attracted. Developing countries also want to make sure they do not lose out from any new system.
Pillar one – reallocation of taxing rights – the new two-tier system
The deal, brokered by the Organisation for Economic Cooperation and Development (OECD), is a two ‘pillar’ solution. Under pillar one, multinational enterprises operating in whatever sector (with a couple of specific exceptions) with a global turnover above €20 billion and profitability above 10% will be subject to tax on a portion of their profits in the countries where they operate. This is meant to catch around 100 of the world’s most profitable multinational groups, but the agreement envisages the threshold being reduced to €10 billion after seven years. If this happens many more groups will be caught.
Countries which will benefit from the new taxing right will be those where the multinational derives at least €1 million in revenue. In order to enable less developed countries to get a share of the revenues, for jurisdictions with GDP lower than €40 billion, the threshold will be set at €250,000.
In exchange for the pillar one taxing rights, part of the deal is that countries, such as the UK and France, which have adopted unilateral digital services taxes, will drop those taxes. However, it is not yet clear when this will actually happen, so a period of double or more taxation across these countries is likely in the short term.
When pillar one was originally devised, it was intended to apply only to digital businesses. This was extended to other consumer facing businesses as the proposals were developed, and now applies to all businesses (other than financial services and extractive industries). It also covers business to business supplies as well as business to consumer supplies. This means there will be more challenges in working out which country is the market for the goods or services and therefore which country gets the taxing rights. For example, if a multinational is supplying goods or services to another business, is the market jurisdiction where the recipient business is based or where the ultimate consumers of the goods or services are based? In the tech field, this could be relevant in multiple areas, such as where cloud services are sold to a business customer which uses them in supplying its customers.
The new taxing right will be over a percentage of what is referred to as ‘Amount A’. This will be the top tier of a company’s profit before tax – any profit in excess of 10% of revenue. Between 20-30% of this will be reallocated to market jurisdictions – the precise percentage has not yet been agreed. It will be divided between market jurisdictions based on their local revenues. However, the amount allocated to a particular country will be capped if marketing and distribution profits are already taxed in that country.
Pillar one will mean that the biggest multinationals – and especially in the tech and telecom sectors - will have some of their profits taxed in jurisdictions where they operate regardless of whether they have a fixed base there. The deal is designed to increase tax take and not merely to reallocate it, so more tax is likely to be paid overall. However, if the new system works, once the inevitable teething issues are sorted out it could bring more certainty than the current position. At present multinationals are subject to a plethora of unilateral digital services taxes (DSTs), which are all slightly different, are based on revenue rather than profits and where there is a significant chance that two jurisdictions will be trying to tax the same profits because of the different approaches of each country with a DST.
However, the detailed proposals for establishing which profits may be taxed where are complex – and not everything has been thrashed out yet, most importantly the hotly-debated revenue-based allocation key which divides up Amount A between market jurisdictions with nexus, and the source rules for specific categories of transactions. Such a major change to the international tax system will inevitably bring a degree of uncertainty and significant added administrative costs for multinationals. As with DSTs, where some multinationals have been open about the fact that costs will be passed onto customers, added tax costs are likely to also feed into increased consumer costs. There will be arrangements to resolve disputes between countries over taxing rights, but whilst the OECD is keen to try to minimise double taxation in principle, inevitably in practice there will be instances of double or more taxation, at least in the short term.
The new taxing rights effectively turn what has hitherto been a tax pie, in which countries fight over their slice under double tax treaties and existing international principles, into a two-tier gateau in which the top tier is the new Amount A split between countries on the new basis. The bottom tier continues to be (mostly) calculated and divided according to existing tax rules, although another aspect of the agreement in relation to pillar one (which could apply to all multinationals, not just those above the 20 billion euro threshold), is a new simplified rule for calculating the attribution of profits within a group for baseline marketing and distribution activities. This is referred to as Amount B.
Whilst the OECD are keen to keep the compliance process as streamlined as possible there is no denying that pillar one, for the largest multinational tech and telecoms groups, signals a sea-change in international tax rules. This will impose a significant administrative burden as well as extra tax bills and (most likely) a decade or so of new tax disputes with market jurisdictions until the rules settle down.
Pillar two – minimum tax rate
Although packaged up with pillar 1, pillar 2 is really a completely separate matter tackling a different problem - the assumption that tax authorities are engaged in a race-to-the-bottom tax rate competition to stimulate investment from offshore.
Pillar two of the agreement tackles this with the Global anti-Base Erosion rules (GloBE) whereby multinationals can be forced to pay a minimum level of tax of at least 15%, regardless of where they are headquartered or the jurisdictions in which they operate. This applies to multinationals with group revenue of more than 750 million euros. However, countries can choose to apply the rules to their headquartered multinationals even if their profits are below this level, so the rules could apply more widely.
Under the GloBE rules countries will not be required to increase their corporate tax rate to at least 15%. But if they do not do so, parent companies in jurisdictions which decide to apply the rules will be subject to a top up tax to the extent that their subsidiaries have paid tax in another jurisdiction at less than 15% or deductions may be denied on payments to group members in low tax jurisdictions.
Although GloBE is in essence a minimum tax rate of 15%, the detail is much more complex. Think of it as a type of controlled foreign company rule but with extra add-ons. Unpackaged, the changes encompass:
the "income inclusion rule" – this pulls income from country A to be taxed again in country B where that income is taxed below the minimum rate (15%) in country A;
the interlinked "undertaxed payments rule" – this catches tax base eroding payments by denying deductions for payments to a related party if that payment is not subject to tax under the income inclusion rule; and
a treaty-based 'subject to tax' rule - this imposes withholding tax in country A where a related party payment is not subject to tax in country B at the minimum rate (confusingly, the minimum rate for these purposes will not be 15%, but instead from 7.5% to 9%).
There will be safe harbours but the detail of these is not yet known.
The rationale for the GloBE is to make low tax jurisdictions less attractive to multinationals. The idea is, if you are going to be taxed at 15% anyway, what is the point in routing group profits through an offshore financial centre, adding cost and complexity to your group structure? For this reason, some low tax jurisdictions, such as Ireland, Estonia and Hungary have not yet signed up to the deal. Ireland has launched a public consultation seeking views on the OECD deal, but is likely to come under pressure, not least from the EU, to toe the line.
Since the whole point of the exercise is to level the playing field, countries that have built their reputation on low tax rates without being out-and-out tax havens may be in a bit of trouble. Whilst they may try to retain current levels of offshore investment with exit taxes and other measures to disincentivise migration, they may struggle to continue to attract new investment at previous levels unless they have other points in their favour, such as a highly skilled local workforce in key areas such as IT or data-analytics.
Even if a country’s usual standard tax rate is above 15%, government incentives such as patent box and other IP regimes (e.g. the UK’s patent box which lowers the corporation tax rate to 10%) could result in the GloBE applying. It is likely that countries which offer tax incentives to encourage IP development will move away from lower tax rates in the direction of other incentives such as R&D tax credits. They will have to be subtle about it, though, because the OECD is keen to make sure such reliefs are not used to circumvent the GloBE rules. Consequently, tech and telecoms multi-national enterprises should be wary of getting too excited about R&D tax credit alternatives as a long term fix to replace low headline tax rates.
For established tech and telecoms sector companies headquartered in places like the UK who are used to applying an existing long and exhausting list of anti-avoidance measures, the new GloBE rules may well turn out to be just one more compliance burden that does not necessarily add that much to their overall tax bill when all is said and done.
For those who have migrated their HQ to places which have relied on very low tax rates, it may take a while for all the glitches to be ironed out, and there may be hay still to be made while the sun shines, but in the long run these changes will have much more impact. For such multinational tech and telecoms companies, pillar two is likely to result in higher tax bills, especially if they have taken advantage of tax structuring using low tax jurisdictions to hold key intellectual property assets, and the cost versus benefit analysis of continuing operations in those countries becomes more acute if there is no other reason to be there. The current agreement was for at least a 15% minimum tax rate, but this was the result of a compromise: the US originally wanted the rate to be 21%, so as well as the rate not being definitely fixed at 15%, there is also a chance of future increases.
The way forward
There is still much detail to be agreed, but signatories to the deal have set an ambitious timeline for conclusion of the negotiations. This includes an October 2021 deadline for finalising the remaining technical work on the two-pillar approach, as well as a plan for effective implementation in 2023.
The deal will also need to be ratified in the US and there are likely to be challenges in getting it through Congress. The US favours pillar two but is more lukewarm about pillar one, which some US politicians see as handing over US tax revenues to other jurisdictions.
Assuming the deal ultimately does get ratified, whilst it could eventually bring welcome certainty, there is likely to be uncertainty and confusion in the short term. Telecoms and tech companies operating multinationally need to take specialist advice on the implications of the proposals in the countries where they operate and/or have key markets once the rules are battened down. Whilst the headline proposals only affect the biggest global companies, some of the changes will impact on those with more modest operations.
