Income taxation at the right place at the right time - An analysis of the need for a virtual permanent establishment

Abstract: In the last couple of years, taxation of the digital economy has become a subject that is on everybody’s mind. This has given rise to attempts from OECD and EU, amongst others, to solve this dilemma. For a company to be taxed in a state it must be a resident in that state or be attributed a permanent establishment that requires that the company has a physical presence in the state. This makes taxing digital companies that conduct business in states a challenge since no physical presence is required. The digital economy consists of different characteristics that defines the way companies do business in the digital market. These are: direct network effects, Indirect network effects, economies of scale, switching costs and lock-in effects, and complementarity. The direct and indirect network effects are about the benefits of users, and the size of the user bases. Economies of scale relate to the marginal cost when it is practically zero. The lock-in and switching costs effects is the effect when users benefit from staying with the same company or would lose by changing company for the service they want. Complementarity is when the user derives more use from using two or more complementary goods together. These characteristics play their role in the single- or double-sided market. The single-sided market is when sellers engage with only one type of customers, and multi-sided markets are the opposite, when there are more than one set of customers acquiring different products and services from a company. In the digital economy there are different types of digital businesses operating. In the thesis, four different models are presented, the multi-sided platform (e.g. Social media), Resellers (e.g. Amazon and Netflix), input-suppliers and vertically integrated firms (e.g. cloud computing services). Due to the effectiveness of digital environments, these kinds of digital companies hold an advantage when generating revenue compared to more traditional businesses. The reliance on technology and algorithms are big in these companies, and some of these models have a big use of huge amounts of data and user generated content. It is hard to pinpoint exactly how much value can be attributed to data and users, but they to have a role to play in the value generation of these digital companies. It is, however, clear that these companies have no need for a physical presence to conduct their business. To solve the issue of taxation of these companies, OECD and other organizations has presented a so called virtual permanent establishment. Instead of relying on a physical presence of the companies, the virtual PE presents other factors to connect the company to the state. The OECD has presented three different factors that may be used when defining the virtual PE; revenue factors, digital factors, and user factors. The revenue factor is based on a revenue threshold. The revenue looked at should only include digital transactions with residential customer through the company’s digital platform. The digital factors try to conclude if the company aims to target residents in its ongoing business, and could be a local website, local payment options or a local domain name. The user-based factor consists of a threshold of monthly active users, online contract conclusions or data collected. Both India and the EU has chosen to implement the revenue factor and user factor in their suggestions for a virtual PE. For attributing profits to the virtual PE, the OECD has presented two “new” methods, the fractional apportionment method and the deemed profits method. The fractional apportionment method basically means that the incomes and expenses connected to the transactions of the virtual PE would be split between the virtual PE and other parts of the company, depending on how much of the income and expense that can be attributed to each part. The deemed profits method is based on a fictional profit, depending on which type of industry the company belongs to. Instead of looking at the real expenses, a presumed expense is used to calculate the income of the virtual PE. The OECD suggests the deemed profits method, since the fractional apportionment method could result in different taxation between traditional and digital companies. However, the EU has concluded that the fractional apportionment method is sufficient and are using a “profit split method”. There is some debate on the real value of data and users, and if these really should be factors that could connect a company to a state. Since the question of digital taxation is best solved internationally, it might be best to use a method that is not as criticized, since we are many states that will have to agree to a solution.