The political pressure to change how hospital governance is currently organized comes from a mix of sources, including increasing costs, technological advances, changing patterns of disease, and growing patient expectations. This pressure to reform has prompted a review of hospital ownership and a drive, in many countries, to decentralize hospital governance. The key trends in decentralizing hospital governance, which may not be compatible with each other, include increasing hospital autonomy and more direct managerial control of hospitals; and decentralization to newly created administrative layers in the health system: –. Hospitals with greater autonomy are perceived to be more flexible in meeting the needs of the local population and, thanks to active purchasing, more efficient performers. –. Decentralizing hospital governance to allow decision making at the local level is seen as a way of promoting responsiveness to communities and to changing patterns of disease; it is also a means of moving responsibility for funding or adopting technological advances away from central government. Despite these apparent merits of decentralized hospital governance, there is an explicit need to: –. acknowledge the tension between potential efficiency gains from centralization and the potential to improve responsiveness and flexibility with greater decentralization. –. address investment decisions (e.g. strategic planning of hospital infrastructure) to ensure geographical equity and efficiency and to balance this with the ‘narrower focus’ decision making in a locally run system. There are good examples of authorities in systems with decentralized hospital governance collaborating to overcome equity and efficiency concerns, but the specific institutional and political contexts make it difficult to transfer successful models from one country to another. More broadly, unless this is supported by the existing country context, policy-makers will struggle to achieve the desired degree of decentralization in hospital governance.