The paper proposes to use regulatory competition theory in order to better understand the evolution of the EU member States’ asylum legislation. It argues that regulatory competition theory can explain the rapid trend of legislative amendments from the mid-80’s onwards, the progressive yet incomplete convergence of the EU member States’ legislation, and the spiral of restrictions of legal norms originally enacted to protect asylum seekers. Competition among legal norms also explains EU Member States’ reticence to collaborate and share the burden. The first argument of the paper is that a phenomenon of competition developed because Member States were convinced that generous asylum policies would be a pull factor for asylum seekers. They feared that regulatory arbitrage (i.e. asylum shopping) would lead asylum seekers to select their destination State on the basis of the level of protection offered. States have entered into a process of de-regulation and, because of their interdependence, national measures have become instruments of a general race to externalise. The result has turned out to be negative and corresponds to a « race-to-the-bottom ». This negative result can be observed at two levels : competition was detrimental to both asylum seekers and States; the rules enacted were suboptimal. The paper then explains why the first cooperation instruments introduced at the end of the 80’s and onwards have failed to meet their objective. The effects of cooperation schemes like the Dublin Convention or burden sharing projects are negligible and there was no shift from costly and unilateral aylum policies towards fairer and more efficient collective action. It is an example of cooperation in the shadow of competition. Finally the paper evaluates the communautarisation of the competence to act in the field of asylum. It is unlikely to permit the emergence of a federal and centralised regulation able to change the nature of the game. In conclusion, the paper seeks to assess if a good combination of cooperation and competition is likely to produce « good » asylum policy in Europe and investigates the “co-opetition” model promoted by Esty and Gerardin.