Should we supplement physical PE with a virtual PE to tax digital services?
Kluwer International Tax Blog
October 24, 2017
Please refer to this post as:, ‘Should we supplement physical PE with a virtual PE to tax digital services?’, Kluwer International Tax Blog, October 24 2017, http://kluwertaxblog.com/2017/10/24/15735/
India’s Income Tax Appellate Tribunal, Bangalore on September 28, 2017, handed down a decision in which it noted that the profits of a foreign company based in Saudi Arabia cannot be taxed in India under the India-Saudi Arabia tax treaty because the company did not have a service permanent establishment (PE) in India. Under Article 5(3)(b) of the India-Saudi Arabia tax treaty, a foreign company is said to have a service PE in the source country if it furnishes services in the source country, including consultancy services, through its employees or other personnel and such services continue for a period of 182 days or more within any twelve-month period. “In our considered opinion,” the Tribunal reasoned, “there is no PE because only ‘solar days’ are to be considered and not ‘man days’ [and] in the present case, the stay in India of the taxpayer was only 90 days and since it is less than 182 days, there is no PE.”
Countries have long debated whether it is appropriate to have a day-counting test in the service PE Article because it could be used to set up artificial arrangements to avoid tax in the source country, and that the duration for which the employees or personnel stay in the source country is irrelevant to the right to tax profits generated from such business activities. The problem arising from the use of a day-counting test as a threshold for the creation of a service PE is quite similar to the traditional physical PE requirement of foreign companies to have a physical or representative presence in the source country. Readers will recollect that in another recent case, a different Bench of the Tribunal opined that it is not necessary for the employees of a foreign company to be physically present in the source country for its profits to be taxed under the service PE Article of the India-UAE tax treaty.
There, the Tribunal noted that a service PE would exist as long as the foreign company is providing services for a certain duration. I have blogged about that case here. These two cases throw two different situations: first, a situation where a foreign company sends its employees or personnel to the source country to render services but the employees are not physically present for the requisite time period; and secondly, the foreign company provides virtual services to its customers located in the source country but does not have any physical or representative presence there. In both these situations, the current service PE rule (and the physical PE rule in some cases) acts as a hindrance for source countries to tax business profits of foreign companies that are generated in the source country through the provision of the services.
Importantly, in both these situations, the foreign company is using the source country’s infrastructure to earn revenue, is integrated into the economic life of the source country, is creating value in the source country, and yet is paying no income tax in the source country. The first situation is easier to tackle. Countries need to reach a political agreement on whether or not to lower the current day-counting threshold. Of course, the line has to be drawn somewhere and it cannot be asserted with full certainty and conviction that one threshold is better than the other. In any event, this is essentially a policy concern and if countries feel that the 90-day or the 182-day threshold must be retained because it signifies a certain degree of involvement of a foreign company with the market country’s economy, then so be it.
Lowering the PE threshold
The main concern is with respect to digital businesses who do not need to be present in the source country (because of the use of digital business models) and can provide virtual services to generate profits. This requires a reconceptualization of the current PE rule to re-allocate taxing rights between the source and residence countries. In Action 1 of the Final Report on BEPS Action 1, the OECD discussed few options to tackle this situation including the option to introduce a new PE nexus based on a “significant economic presence.” The EU is currently debating on this too, with Austria being the first country to make public its intent to enter into bilateral treaty negotiations with Ireland with a view to incorporating this new nexus.
Whether and how such re-allocation needs to be done requires careful consideration of existing international tax principles (neutrality, efficiency, certainty, and simplicity) so that these basic principles governing the PE rule are kept intact. There is no doubt about the fact that the PE rule has, traditionally, played a significant role in allocating taxing rights between source and residence countries based on the sourcing and benefit theories and a deviation from the current PE threshold must not go unquestioned. But it is equally true that noted international tax scholars such as Dale Pinto and Arvid Skaar have long pointed out that the traditional PE rule is not so “sacred” that it cannot be redesigned to adapt to changing times (read, the rise of digital economy).
Is it not true that the PE concept has been diluted several times, from introduction of a construction clause to a server or a service PE, to the recent revisions to the dependent agent PE clause? Is the current PE rule which places an excessive reliance on physical presence suitable for digital business that can virtually participate in the economy of the source country through the use of technology? Is the source country not justified to tax profits from provision of services on the basis that it provided an opportunity to foreign digital companies to use its infrastructure, including its stable legal and economic system, and to generate profits in the first place, and most importantly, offered protection of intellectual property rights? These are some of the tough questions that the OECD’s Final Report on BEPS Action 1 conveniently ignores.
In its Report, the OECD discussed the idea of creating a new PE nexus but did not recommend it at this stage for a lack of political consensus. The OECD noted that the digital economy is becoming part of the economy and therefore it would be difficult to ring fence digital economy and the introduction of a new PE nexus would imply deviation from long-standing principles of international tax law governing allocation of taxing rights between source and residence countries. The phrase “significant economic presence” is not easy to define and the OECD has also warned that specific factors may not accurately determine “significant economic presence” of a foreign company because these factors can be easily influenced and therefore the countries may use a combination of specific factors to determine whether a foreign company can be said to have a “purposeful and sustained interaction with the economy of the country concerned.”
A consensus-based approach
There are practical challenges too. Experts question how a new PE nexus can be measured and determined and, if agreement is reached on the model and design (say, using the number of active users a foreign company has in the market country, or by using a de minimis revenue threshold), what use can the nexus serve if the profits cannot be properly attributed to it? Some suggest that the challenges at hand can best be tackled under the consumption tax regime. It is anybody’s guess what the OECD’s Final Report (due in 2020) on digital economy taxation would look like, but the idea for countries is to refrain from taking unilateral measures in domestic tax laws to avoid potential breach of treaty obligations. If need for a quick fix appears urgent, the best way forward for countries would be to work collectively and adopt a consensus-based approach to taxing the digital economy.