Decentralization and health performance in Italy: theoretical and empirical issues

Three chapters constitute the main structure of this contribution. The first chapter investigates the relationship between fiscal decentralization and regional health outcomes, as measured by infant mortality rates, in Italy. The paper employs a panel of all Italian regions over a period of 17 years (from 1996 to 2012), applying a linear Fixed-Effect model. Two different quantitative measures of fiscal decentralization are used, which capture the degree of regional decision-making autonomy in the allocation of tax revenues and the extent of regional transfer dependency from the central government (i.e vertical fiscal imbalance). Methodologically, to account for the temporal dynamics of the decentralization impact, the robustness of the findings is checked, among others, with respect to the use of an Error Correction Model, which allows to disentangle short and long run effects. The analysis also deals with the issue of heterogeneous distributional geographical responses by modelling the asymmetric impact of decentralization on infant mortality rates according to the level of regional wealth. The second chapter addresses the issue of convergence. Here the main research questions are whether health outcomes in Italy converge/diverge over time and, more importantly, whether decentralization has played a somewhat role in the convergence/divergence process. Using a pooled dataset with the same time span as the previous one, the conventional measures of sigma- and beta- (both absolute and conditional) convergence are estimated for two different regional health outcomes (i.e. infant mortality rate and life expectancy at birth). Again, two measures of decentralization are employed in order to catch both the degree of fiscal regional decision-making autonomy (i.e. the same indicator as in chapter 1) and the political decentralization dynamics (i.e. a dummy variable taking the value of 1 after the introduction of the 2001 constitutional reform). From a methodological point of view, the real novelty of the analysis is to take spatial dependence and neighbourhood effects among the regions into consideration. Modelling the impact of decentralization through an interaction term, the speed of convergence is found to be significantly affected by the level of decentralization. The third and last chapter deals with the issue of the effects of decentralization from a different but related viewpoint. Compared to the previous chapters, it examines descriptively the administrative aspects of decentralization by a lower (micro) level perspective, looking at the managerial autonomy of local healthcare structures. The emphasis is here on the role of intrinsic and extrinsic motivations in enhancing accountability and improving the performance of healthcare system, in general, and the quality of hospital care, in particular. Though the focus is not specifically on the Italian system, the analysis is particularly relevant for this country, where regional governments, in charge of the responsibilities for the financing and the delivery of healthcare, act through a network of Local Health Authorities i.e. public entities with their own budgets and management, which directly run small public hospitals -, public hospital trusts with full managerial autonomy and accredited for-profit private providers. The understanding research hypothesis here is that the way in which the financial incentive schemes for providers are designed and structured is likely to affect their effectiveness in pursuing the expected results (e.g. improved efficiency and quality of healthcare service delivery). However, the same incentive is expected to work differently according to the provider s degree of decision-making autonomy and its utility function.