This paper synthesises the findings of the RISE Political Economy of Adoption (PET-A) country studies by interpreting their findings in light of the findings of comparative research on the evolution of ‘welfare regimes’. This paper suggests that PET-A countries have not, for the most part, developed the sorts of political settlements identified as conducive to change in the welfare regimes literature or indeed others that might be expected to produce similar gains. They are instead dominated by predatory coalitions which seek to use education systems for rent-seeking, ideological, or other non-developmental purposes while technocratic and progressive elements are marginalised. In all cases, they consequently lack coalitions that are capable of driving education development. At the same time, the PET-A studies also provide some evidence to suggest that democratic reform can trigger improvements in education policy and learning outcomes by shifting the balance of power between elements within political settlements. In this respect, the PET-A studies suggest that the key to promoting educational development/learning in the developing world may be to think beyond narrow education-focused interventions to broader programs of democratic rights-oriented reform.In presenting this argument, this paper begins by providing an overview of the welfare regimes literature’s findings with regards to the political dynamics that have facilitated education and learning gains. It then assesses the extent to which the PET-A countries have developed such political settlements or indeed any others that might produce learning gains and reviews the evidence in the PET-A studies as to the effects of democratic reform. The final section of the paper presents the conclusions.