Reforming decentralized integrated health care systems: Theory and the case of the Norwegian reform

In this essay a conceptual and theoretical scheme for decentralized integrated health care systems of the northern European kind is developed. With small changes it is also applicable to other countries, e.g. Italy, Spain, and Portugal. Three ideas tie together the scheme: modified fiscal federalism, principalagent thinking and the analysis of discrete structural alternatives from new institutional economics. As a special case it encompasses the ideas of planned markets and public competition developed by von Otter and Saltman. The scheme can be used to analyse driving forces behind reforms and prediction of effects. To illustrate the thinking the recent Norwegian reform is put into context, not only geographically but also theoretically. The geographical context is that of Scandinavia and there is a summary of reforms in the Scandinavian countries over the past 20-30 years. The essay thus serves the double purpose of presenting and evaluating the Norwegian reform in a Scandinavian context and to take part in the neglected discipline of developing a theory of health care reform. The Norwegian January 2002 reform is described in some detail. It is a reversal of the Scandinavian model of decentralization and a move towards more centralism. The hospital system was transferred to the state that established five regions with independent (non-political) boards and each region has a number of daughters (hospitals) that have great autonomy with their own boards and are outside the legal restrictions of the public sector. Basically the idea is to mimic the corporate structure of large private companies. The reform is evaluated based on principal-agent thinking and the analysis of discrete structural alternatives. Overall there is no a priori reason to expect large improvements in efficiency – but on the other hand neither should one expect things to get worse. Many effects depend, however, crucially, on (a) the financing system that will be put in place late 2002 or early 2003, and (b) whether or not the political and management culture change as a result of the reform. In the concluding sections possible implications for Denmark and Sweden are discussed.