Università degli Studi di Cagliari. Cagliari (Italia)
30 de marzo de 2009
Resumen:
This PhD dissertation contains three essays within the area of regional economic modelling and, although each essay can be seen as an independent manuscript, they are linked to each other. The main purpose of the dissertation is to study regional development policy and regional adjustment mechanisms with the help of applied general equilibrium models. The first essay describes the building of a Social Accounting Matrix (SAM) for Sardinia which provides the data set necessary for calibrating a general equilibrium model. Scarcity of information at a regional level, means that a well-defined SAM cannot be created using a simple compilation method. For instance, the lack of data on secondary distribution processs inhibits an accurate compilation of the sub matrix of transfers between institutions. This sub matrix is estimated by means of a doubly constrained minimum information (MI) model, originally developed by Plane (1982) and Schneider and Zenios (1989). In order to take account of measurement errors, I have introduced random noise into the MI model using the same method as Robinson et al. (2006) in a Cross Entropy (CE) model. Furthermore, the CE model is used to reconcile all the information employed and to introduce additional precisions in the estimation of some macro aggregates. The role played by knowledge capital as a factor of regional development is the topic of my second essay, “R&D Investment and External Knowledge Spillovers”. I present the model for Sardinia with an analysis of the macroeconomic consequences of policy promoting R&D. The model is a single-county dynamic general equilibrium model where agents have myopic behaviour. Contrary to other regional myopic models found in the literature, this model for Sardinia incorporates labour market imperfections and allows for total adjustment of labour and capital stock through migration and adjustment cost functions. The third paper is titled “Does the Forward Looking Model Fit the Regional Economic Features?”. Intertemporal forward looking models are usually calibrated on national data, however a slavish application of the characteristics of such models may cause some problems in a regional context since regions may differ from the country as a whole. It is argued that intertemporal consumers’ optimization based on neoclassical or Fisherian intertemporal resources allocation is inappropriate to the region since endogenizing the path of savings involves a balance of payments constraint, a constraint not faced by regions. Furthermore, I compare forward looking and myopic models. This comparison may be very useful since, in the literature, the intertemporal model has generally been compared to a simple static case lacking any capital adjustment rule. Contrary to previous exercises, we find that the only difference between the two models is in the transitional pathway where consumption and investment might diverge since agents with perfect foresight have ii rational expectations, whilst those with myopic foresight take decisions according to adaptive expectations without making any intertemporal preferences between periods on future profit and income.
Cita:
P. Lecca (2009), Three Essays on Regional Economic Modelling. Cagliari (Italia).
