• Pillar One – revisiting earlier predictions

    B.M.R. On

    So how have my earlier Pillar One predictions (in this post here) turned out so far?

    First, I had predicted that the mid-2021 target date (for securing Inclusive Framework agreement) would be missed.

    Well, we don’t have unanimous agreement of all Inclusive Framework members, so we can safely say the target has been missed. Most members are on board, but there are some notable holdouts – for Africa, that would be Nigeria and Kenya.

    Next, I had predicted an increase, during 2021, of digital tax measures, to be introduced by various countries, as they wait for the global solution to take wings.

    This prediction has come to pass. In fact, even now in 2022 (even after the publication of more concrete proposals from the Inclusive Framework), such unilateral measures are still being introduced. For example, Nigeria has recently announced a 6% digital services tax. Perhaps this particular move should come as no surprise, as Nigeria had already indicated its dissatisfaction with the Pillar One outcomes, and declined to sign the agreement. That said, it’s a bit strange to see this new tax, given that Nigeria had only recently (in 2020) introduced a significant economic presence tax regime, ostensibly targeting (among others) revenues from digital businesses.

    I had also predicted that the slew of unilateral taxes would lead to the introduction of standard tax treaty provisions providing relief for double taxation, and that, this ‘organic’ approach ‘will render Pillar 1 redundant’.

    This prediction has not (yet) come to pass, but I still believe that things will pan out this way. There’s an added complication, though. In the months since my post was published, the scope of Pillar One has been expanded beyond digital businesses. As a result, it no longer focuses solely on high revenue-yielding digital businesses. As such, the risk of double (or multiple) taxation has increased in scope, albeit for different reasons. However, focusing strictly on digital businesses, I still believe that we will see tax treaty solutions begin to emerge.

    I had also stated that I would be ‘very surprised if Pillar 1 ever saw the light of day as a concrete plan’.

    I remain of this view. Despite its adoption by most members of the Inclusive Framework, there remain misgivings about certain key aspects of the proposals. For one thing, we are far from clear on all the technical aspects of Amount B, an issue of particular relevance for developing countries. There is still some way to go.

    Overall, I stand by my earlier predictions. Let’s keep watching, see how things go.

  • Pillar One – a prediction

    B.M.R. On

    The G20 Riyadh Summit concluded today. On international taxation, there was (as expected) broad support for the work of the OECD Inclusive Framework.

    The G20 Leaders expect that the work will be concluded (as earlier promised by the OECD) in mid-2021.

    Well, I don’t think this will be achieved. The outstanding technical issues are a big obstacle, moreso as any agreement would require the unanimous approval of the members of the Inclusive Framework.

    As far as digital taxes are concerned, here is my prediction: over the next few months, there will be an increase in unilateral measures to tax digital businesses. Countries that have held back up till now will begin to introduce these new taxes. While many countries had held back, waiting instead for a ‘global solution’ from the OECD Inclusive Framework, there is an understandable impatience in the air. Countries don’t want to miss out on tax revenues, and the recent delay (by the Inclusive Framework) is already prompting some countries to introduce their own measures. In this, they would be joining those other countries that had earlier on decided not to sit back and wait for a global solution, but rather to introduce their own rules in the meantime.

    And this will, of course, lead to double, and even multiple, taxation across jurisdictions. Eventually, countries will negotiate bilateral agreements to reduce or eliminate this double taxation.

    This is most likely how the digital taxes issue will be resolved. We will also likely see common features emerge in the various unilateral digital taxes being introduced, making it easier for there to be standard treaty provisions for double tax relief.

    This ‘organic’ approach will render Pillar 1 redundant. The combination of unilateral (taxing) measures and treaty provisions (for double tax relief) will see to that.

    I think this is the way that things will go. I would be very surprised if Pillar 1 ever saw the light of day as a concrete plan.

    And as for my prediction for Pillar 2, I’ll leave that for another blog post.

  • Article 12B gets green light

    B.M.R. On

    The UN Tax Committee has reached a decision: the proposed Article 12B (on the taxation of income from automated digital services) will be included in the UN Model Tax Convention.

    I remain doubtful whether this provision will actually make it into any real-life tax treaties, especially those treaties that matter where such income is concerned. For example, the United States (typically the residence state for the largest digital businesses) is unlikely ever to agree to such a provision in its tax treaties.

    Also, the provision can only have proper effect if the treaty partners actually have domestic legislation taxing this income. The treaty itself cannot give a right to tax if such income is not taxable in the jurisdiction of the treaty partners.

    And even if the treaty partners do have domestic laws taxing ‘digital income’, it depends further on the particular type of tax regime in place. This Model treaty provision applies only to ‘payments’ for automated digital services. As such, it won’t affect taxing rights that apply where a payment has not been made. For example, where the country levies a tax on the ‘value created’ (within its jurisdiction) by the non-resident company, even if there has been no actual ‘payment’ (from its jurisdiction) to that company.

    Still, considering the slow progress over at the OECD, one must not quibble over this development. It’s a start. Although one does rather wonder at the point of including (in a Model treaty) a provision that has scant likelihood of ever being used.

  • Draft Article 12B – ATAF’s viewpoint

    B.M.R. On

    ATAF has published a technical review of the proposed Article 12B of the UN Model.

    Interesting because, if more African countries introduce digital services taxes (and ATAF recommends that they do), then these will have to work in tandem with any applicable treaty provision on automated digital services.

    And, as the UN Model draft provision is the only such treaty provision on the table at the moment (albeit still in draft form), it’s worth the extra attention it’s getting from all quarters.

    Several points in the ATAF document, but this one stood out for me: ATAF cautions that the scope of the UN Model draft provision is much narrower than what ATAF has been recommending for African countries. The proposed Article 12B applies only where there is a payment from the source country to the residence country (i.e. the country where the automated service provider is resident).

    ATAF points out that, where the automated digital service is supplied free to a user in a market jurisdiction, the proposed Article 12B would not apply. This could happen, for example, where the service is provided for free to users within a particular jurisdiction, and the payment (to the service provider) comes in the form of advertising revenue, for example, from a third party in another jurisdiction.

    The proposed Article 12B concerns itself only with the existence of a payment for automated digital services, seeks to identify the jurisdiction where that payment is sourced, and allocates the taxing right between that (source) jurisdiction, and the jurisdiction where the service provider is resident.

    However, ATAF recommends a much broader approach. It recommends that countries levy digital services taxes based on ‘user participation’. Under this approach, a country would levy the tax if there were users of the service within its jurisdiction, even if those users are not themselves paying for the service. ATAF points to the value (to the service provider) created by user participation, together with the attendant network effects. ATAF insists that countries should be able to tax such value streams, irrespective of whether the revenues come from the users themselves.

    Solid point, but again, that brings us to the issue of implementation. Levying a tax on the basis of user participation would be horrendously difficult. Those complexities have already been seen in more developed countries (see, for example, the discussions around the UK Digital Services Tax). We can expect the Africa situation to be even more challenging. Against that backdrop, the draft Article 12B approach appears a more realistic way.

    You can access the ATAF document here.

  • OECD Tax Talks – what next?

    B.M.R. On

    Yesterday, the OECD gave us an update on the work of the Inclusive Framework on Pillars I and II. No agreement on the final package, but here’s what we got, anyway …

    We got the official reports on the Pillar I and Pillar II Blueprints (see here and here). Much of the content had been heavily leaked beforehand, but there you go.

    We also received the impact assessment. This time around, it includes full details of the methodology used. The previous one had been criticized for not including such details.

    And we also get a chance to comment on the Blueprints – public consultation opened yesterday, and runs till 14 December 2020. Here is the condoc.

    So no agreement; didn’t stop the OECD trying to put a positive spin on things, though – lack of agreement not withstanding, the Blueprint provides a ‘solid foundation for a future agreement’. I suppose we can take that.

    The Inclusive Framework now has a few more months to secure that elusive agreement. The new target is mid-2021.

    I’m still not banking on an agreement being achieved. Rather I foresee a proliferation in unilateral measures across countries. Expect more digital services taxes. This could then lead to a harmonization effort across the board, and perhaps also a treaty solution for allocating taxing rights, and relieving double taxation. A role perhaps for the recently proposed Article 12B of the UN Model? Interesting if things actually turn out that way.

  • A note on Nigeria’s Significant Economic Presence rules

    B.M.R. On

    Earlier this year, Nigeria introduced measures to tax foreign digital businesses that generate income within the country. The new taxing right rests on the concept of a business having a ‘significant economic presence’ within Nigeria.

    The rules took effect on 3 February 2020. They cover both digital and non-digital businesses, however, this post focuses on the digital businesses aspect of the rules.

    ‘Significant economic presence’

    There are three scenarios under which a taxpayer would be deemed to have a ‘significant economic presence’ (SEP) in Nigeria.

    Scenario 1.

    The taxpayer has a gross turnover or income of more than NGN 25 million (around USD 60,000) from any or a combination of the following activities:

    • streaming / downloading services of digital content;
    • transmission of data collected from Nigerian users, generated via a digital interface;
    • provision of certain goods and services via a digital platform to Nigeria; and
    • provision of intermediation services via a digital interface linking suppliers and customers in Nigeria.

    It’s debatable whether the gross turnover threshold has been set at an appropriate level. If the aim is to target only the large multinationals, one would argue that the threshold is rather low, and would also catch other, smaller, businesses. On the other hand, one should also consider the size of the market. While USD 60,000 might not seem like a lot of money for one country, it’s worth setting that amount into the African context. But also worth noting that Nigeria is one of the larger African markets. The threshold doesn’t seem to fit right, in any case.

    Scenario 2.

    The use, by the taxpayer, of Nigeria domain name (.ng) or registration of a website in Nigeria.

    Scenario 3.

    Where the taxpayer has a “purposeful and sustained interaction with persons in Nigeria” by customising its digital page or platform to target persons in Nigeria (localised pricing, billing, payment).

    Scenarios 2 and 3 do not contain a revenue threshold. This means that even small businesses can be caught by the rules – for example, a non-resident SME that supplies digital services to Nigerian customers via a Nigerian website, and using localized pricing. Here the amount of turnover is immaterial – the turnover condition applies only to Scenario 1.

    What happens if / when a global solution is found?

    The OECD has been doing its utmost to persuade countries to hold off on introducing unilateral measures to tax digital businesses, and instead to wait for a global solution from the OECD Inclusive Framework. However it also realises that many countries will plough on regardless.

    Bearing that in mind, the OECD issued some guidance for countries that wish to introduce unilateral measures.

    The OECD advises that such unilateral measures should have the following characteristics:

    • they should be temporary;
    • they should be targeted;
    • they should minimize overtaxation; and
    • they should have little to no impact on small companies.

    So how do the Nigerian rules fare?

    Temporary?

    Well, not exactly. The rules provide that they will remain in place until a global solution (i.e. an international agreement) is in place. But they will only be withdrawn in respect of any taxpayers or transactions that come within the ambit of the international agreement. For those falling outside the agreement, the rules will remain. This is quite a sensible belt-and-braces approach, even though it will likely lead to complexity. Worth contrasting that with the position in other countries (for example, the United Kingdom) where such taxes have been introduced with the clear commitment that they would be withdrawn once a global solution is in place. The Nigeria position is a kind of halfway house.

    Targeted?

    To some extent. As mentioned above, the rules cover both digital and non-digital businesses. As far as the digital part is concerned, it is clear the types of businesses being targeted. The issue though is that the net might be being cast too wide.

    Which leads on to the next point.

    Minimize overtaxation?

    Absolutely not. As stated above, the threshold in Scenario 1 might be on the low side. And, as for Scenarios 2 and 3, there is no threshold at all. As drafted, the provisions would land heavily on small businesses.

    This also links in to the point about the impact on small businesses.

    Limitation of the SEP rules

    There is one serious limit to the rules, though.

    They are domestic rules, and therefore will be overridden by any of Nigeria’s tax treaties that contain a permanent establishment provision. That provision will prevail over the domestic law rules, with the effect that the relevant income will only be taxed in the source state (in this case, Nigeria) if the non-resident company has a permanent establishment in Nigeria. As the standard definition of ‘permanent establishment’ is not wide enough to cover the Nigerian ‘significant economic presence’ concept, that concept will be overridden by the treaty concept of permanent establishment.

    The SEP rules will therefore work only as concerns a jurisdiction with which Nigeria does not have a double tax treaty.