Member states of the European Union (EU) respond differently when they get caught for violations of European law and face prosecution. While Portugal tends to settle its non-compliance cases quickly and at an early stage of the EU's official infringement proceedings, Italy and Belgium like to sit them out and do not even comply with rulings of the European Court of Justice after being convicted twice – first for violating EU law and then for not acting upon the court's original judgment. This paper explores the explanatory power of prominent compliance theories to address these diverging patterns of persistent non-compliance across EU member states and tests a set of hypotheses that accounts for the non-compliance dynamics across the different stages of the EU's infringement proceedings and over time. We find that both differences in capacity and power explain the variation in longitudinal non-compliance patterns.