The project deals with the comprehensiveness, coherence and clarity of the Supreme Administrative Court's (SAC) practice in tax cases, including the application of tax treaties. The aim is to present an updated tax law interpretation methodology. There are examples of HFD decisions that may indicate a lack of coherence in how choices are made between different interpretation objects/sources of law, especially in difficult cases. This creates weak predictability and uncertainty in the application of law. The starting point for the project is that tax law decisions should be created through rational argumentation based on given objects of interpretation and with the support of clearly stated prioritization principles in as uniform a manner as possible, namely according to what Strömholm calls principle-following judging. Leave to appeal is granted in only about 2% of the judgments and decisions appealed from the courts of appeal. In practice, the administrative courts of appeal are the last instance in almost all cases and the role of the Supreme Administrative Court is primarily to set precedents. It is important that the courts of appeal are guided by careful precedent setting by the Supreme Court. This requires that the Supreme Administrative Court gives detailed reasons for its judgments and issues judgments that are coherent with each other. The research will identify and analyze possible incoherence. If there are justified reasons for the lack of coherence, these will be evaluated and the reasons for them weighed against the reasons for principled judging.