Four Essays on the Productivity Convergence of the European Regions

The thesis aims to empirically verify if it exists and what features assume the process of regional productivity convergence at the European level. In doing this, we pay particular attention to the role of spatial effects and European cohesion policy. Under a methodological point of view, we use econometric models capable of handling the phenomena of spatial dependence, allowing us to obtain reliable estimates. Moreover, the effect of Cohesion policy is evaluated through the introduction of variables relating to regional expenditure for the Structural Funds. The PhD thesis is organized in four chapters. The first chapter investigates the total and sectoral convergence of labour productivity between NUTS-3 regions of EU-12 over the period 1980-2010. A β and σ-convergence approach is adopted along with a methodology based on Getis’ spatial filters that allows decomposing each variable into its spatial and a-spatial components ensuring the spatial independence. The estimates highlight a process of σ and β-convergence in regional economies. Furthermore, it is observed that i) the employees expelled from agricultural sector are not fully absorbed by services and manufactory sectors, ii) the productivity of services and manufactory sectors is increasing in lagging regions with the consequent decreasing variance, and iii) the β-convergence process is in act within sectors and for the overall productivity. Finally, it has been found that iv) higher productivity services are rather concentrated in some cities and regions. The second chapter aims to evaluate, by means of cross-sectional growth modelling and spatial econometric analysis, whether Structural Funds were effective for regional convergence of labour productivity among regions of the European Union from 1989 to 2006, and what factors affect the effectiveness. In the β-convergence model, beside Structural Funds and conditioning variables, it has been included an interaction term comprised of regional institutional quality in conjunction with provision of Structural Funds, in order to evaluate their joint returns on growth. The effects of Structural Funds upon convergence are assessed through a spatial filtering technique that manages both spatial dependence in the residuals and the possible existence of multiple steady states. This approach, although it is an economic a-theoretical method, leads to a generalization of the Solow growth model in which each region follows this model, but their aggregate production functions and their steady states are free to vary. Three main conclusions have been reached. First, Structural Funds for Objective 1 help to accelerate the process of convergence. Second, the interaction of regional institutional quality and Structural Funds for Objective 1 has a significant and positive effect upon regional productivity growth. Third, European regions have varying paths of growth, and in this study clusters of regions with homogeneous convergence rates have been delineated. In the third chapter, the structural and within components of productivity growth are examined. The analysis starts from the idea that regions that manage to improve their living standards need to be able to diversify away from traditional sectors in order to focus in more advanced. The speed at which this diversification takes place determines the success, or failure, of a region. In the first part of the chapter, an exploratory spatial analysis leads to observe that the structural and within variation are strongly geographically clustered. The estimation of the factors that affect the within and structural change is then performed through a Spatial Durbin econometric model able to explicitly include the spatial spillover effects. Empirical estimation shows that the structural change is promoted by market potential via spatial spillovers and it is negatively affected by high regional specialisation, diseconomy of scale and heterogeneous local conditions. The initial presence of low productivity sectors as well as the agglomeration economies do not have any impact and the Structural Funds, along with institutions, have a role in promoting the structural variation. Finally, the structural change is a key factor in order to cushion the fall of productivity between 2008 and 2009. The fourth chapter examines the impact of the Structural Funds on the productivity growth of the Spanish regions over the period 1989-2006 adopting a Spatial Durbin panel model. Our main findings regard the mixed impact of Structural Funds for Objective 1. The resources devoted to business and agricultural support have a positive impact on productivity. At the opposite, the resources for infrastructures play a negative role on productivity, probably because they accentuate the weaknesses of the regions with low endowment. The same negative impact is produced from Funds on human capital since these do not translate into an effective productivity growth. Finally, the effects of Structural Funds tend to vanish soon over the time.