Skatterättslig tolkning i ljuset av BEPS-åtgärderna

Sammanfattning: One of the most effective ways for states to raise revenue for their treasury is through taxation. This is even if the value creation takes place outside the country's borders. The problem is usually referred to as double taxation. As a consequence of this, from a micro perspective, uncertainty arises for taxpayers in cross-border transactions. From a macro perspective, potential investments by companies such as expansions to other countries are also endangered. One of the solutions to this problem has been for the states to enter into so-called double taxation treaties with each other, which regulate, amongst other thing, which state has the right to tax a certain income generated by a taxpayer. Since the 1960s, the OECD has developed a model convention that member states can use when drafting their tax treaties. Commentaries have also been developed for the articles of the model agreement which facilitate the interpretation of the tax treaty. The model agreement and comments are regularly revised. Sweden is one of the original member states of the OECD and a majority of the Swedish tax treaties are based on the model convention. The model convention and the legal status of the commentaries have been the subject of constant legal discussion over the years based on international tax law. Especially with regards to how states' tax authorities and courts should treat them when interpreting tax treaties. Recently, a new problem has also arisen for the states, namely double nontaxation. This means that a cross-border situation is not subject to any taxation in any of the contracting states. The problem in this regard is consequently that the tax bases of the states are being eroded and economic relations are deteriorating between the states. The OECD launched the BEPS-project (Base-Erosion and Profit Shifting) in order to counter this problem, which resulted in 15 actions. In terms of double taxation treaties, action 15 is the most relevant, named for the Multilateral Instrument (MLI). The MLI is a convention whose purpose is to enable the states to make modifications in their concluded tax agreements without having to call for bilateral negotiations. The question that now arises is how the interpretation of the Swedish tax treaties may be affected due to the fact that Sweden will implement MLI on the tax treaties that are to be covered by the convention. The study therefore intends to investigate this particular matter. This is done by first giving an account of how tax treaties have historically been interpreted by the Supreme Administrative Court. Afterwards, an account will be made of the relevant articles of the MLI and the BEPS-project that Sweden has chosen to incorporate into their internal legislation. The study concludes that the OECD model convention and its commentaries have had a significant impact on the interpretation of tax treaties from a pragmatic perspective. From the legal perspective, there is still uncertainty about their legal status, particularly with regards to the Vienna Convention from 1969 on the law of treaties, whose articles of interpretation have been explained in the study. The legal cases referred to in the study span over five decades in order to identify patterns regarding the interpretation of tax treaties using the model convention and its commentaries. The clearest pattern is that the Supreme Administrative Court makes use of the comments in cases where the relevant provisions of the tax agreement are based on the model agreement. However, it is not quite clear how the court has reasoned about the use of a particular version of the commentaries. In addition, the study has also concluded that the preparatory work for the tax agreements plays a large role when it comes to determining the common intention of the parties, which the Supreme Administrative Court has attached decisive importance to in the interpretation of tax agreements, regardless of whether or not they are designed in accordance with OECD model convention. Regarding MLI, the study concludes that the impact on the Swedish tax agreements is relatively small due to the fact that Sweden has only chosen to introduce the articles from the convention that constitute the minimum standard. Of these articles, it is only the article that stipulates that the covered tax agreements must include a so-called Principal Purpose Test (PPT) which may possibly lead to increased complexity for tax authorities and courts in the application of the article, as well as unpredictability for taxpayers. In this regard, the work concludes that the model convention of 2017 and its commentaries (as well as future revisions) may play an important role in the interpretation of the PPT. Furthermore, in the study, an analysis has also been made with regards to how Sweden chose to implement MLI in its first tax agreement and how this approach may affect the main purpose of MLI. In this respect, the study concludes that the chosen way that Sweden, together with Great Britain and Northern Ireland, chose to implement MLI in its tax agreement (through an amendment protocol) is satisfactory mainly on two grounds. One is that it contributes to more certainty for taxpayers. The second is that it facilitates the interpretation of the tax agreement based on the determination of the common intention of the parties, which the Supreme Administrative Court values the most in the interpretation of tax agreements. The disadvantage, however, is that an amendment protocol is to be considered a bilateral negotiation, which MLI intends to avoid. The study concludes that future precedents are required to provide a clear conclusion regarding the interaction that the MLI may have with the OECD model agreement and its associated commentaries, even if the latest version has been designed within the framework of the BEPS-project.