Background: The Malawi government in 2002 embarked on an innovative health care financing mechanism called Service Level Agreement (SLA) with Christian Health Association of Malawi (CHAM) institutions that are located in areas where people with low incomes reside. The rationale of SLA was to increase access, equity and quality of health care services as well as to reduce the financial burden of health expenditure faced by poor and rural communities. This thesis evaluates the implications of SLA contracting out mechanism on access, utilization and financial risk protection, and determines factors that might have affected the performance of SLAs in relation to their objectives. Methods: The study adopted a triangulation approach using qualitative and quantitative methods and case studies to investigate the implications of contracting out in Malawi. Data sources included documentary review, in-depth, semi-structured interviews and questionnaire survey. The principal agent model guided the conceptual framework of the study. Results: We find positive impact on overall access to health care services, qualitative evidence of perverse incentives for both parties to the contracting out programme and that some intended beneficiaries are still exposed to financial risk. Conclusion: An important conclusion of this study is that contracting out has succeeded in improving access to maternal and child health care as well as provided financial risk protection associated with out of pocket expenditure. However, despite this improvement in access and reduction in financial risk, we observe little evidence of meaningful improvement in quality and efficiency, perhaps because SLA focused on demand side factors, and paid little attention to supply factors: resources, materials and infrastructure continued to be inadequate.

Policy transfer defies the notion of national boundaries in policy making and development. With globalisation processes in the picture, purely state centric policy making models are not the only option. International and domestic policy entrepreneurs have been pivotal in transfer processes. For developing countries, international donor organisations have been instrumental through conditioning assistance to policy reform. Due to the prevailing hierarchical aid regimes, the assumption is that developing countries would implement these policies for the fear of losing the much-needed aid. However, this study argues that the actual implementation of reforms emanating from the global arena is not an automatic process even in the context of coercive transfers, as it is mediated by country specific contextual frameworks. Moreover, even if implemented, the extent to which the transfers attain the promised transformation ends depends on prevailing environmental factors, appropriateness of the reforms, and the implementers' in-depth understanding of the reform instruments. The analysis used the cases of hospital autonomy and district health management decentralisation reforms which are based on the new public management (NPM) paradigm to examine the mechanisms of policy transfer; factors constraining or facilitating the adoption and/or implementation of transferred policies; and the impacts of the policy programmes on service delivery transformation in Malawi. A multilevel framework was used to analyse the dynamics at international, national and application levels. It used a qualitative research strategy. Therefore, data was collected through in-depth interviews, focus group discussions, documents, and observations. The study finds that due to Malawi's heavy aid reliance, international donor organisations attempted to introduce the hospital autonomy and district health management decentralisation reforms on its policy agenda through aid conditioning mechanism which has coercive attributes. In the former, USAID as an international institutional entrepreneur was the driving force through its non-project assistance (NPA) aid regime while in the later case it was the European Commission within the institutional framework of the Lome IV Convention. A comparative analysis of the two reforms revealed that a combination of contextual issues of: mode of transfer, policy content and political-economic context, path dependence, parliament-cabinet configuration, bureaucratic politics, pressure from citizens, institutional compatibility and prerequisites, and social economic forces; determined their adoption and subsequent implementation. While hospital autonomy was rejected by cabinet, and not implemented, despite large amounts of donor resources invested in the transfer processes because of these contextual issues, decentralisation was implemented as the environment was favourable, although it met bureaucratic resistance. However, the study found that when implemented, decentralisation faced several contextual challenges including modest levels of application, reproductions, reversals, cultural factors, and unintended consequences so that it has not achieved the intended transformational results. To this end, the findings provide a better understanding of the dynamics of policy transfer in developing countries and work as a springboard for donor organisations to reorient their approach in aiding policy development in developing countries.