LAND QUALITY, DEVELOPMENT AND SPACE: DOES SCALE MATTER?

This study analyzes empirically the relationship between land quality decline and the spatial distribution of per capita income observed in Italy at different spatial scales and geographical divisions. The aim of this contribution is to verify if a decline in land quality has higher probability to occur in economically disadvantaged areas and if scale may influence this relationship. Per capita income was considered a proxy indicator for the level of socio-economic development and life quality in the investigated area. Changes over time (1990-2000) of a composite index of land quality and per capita income in Italy were regressed at four spatial scales: (i) 20 NUTS-2 regions, (ii) 103 NUTS-3 prefectures, (iii) 784 local districts designed as Local Labour Market Areas (LLMAs), and (iv) 8,101 LAU-1 municipalities. Different specifications were tested, including first, second and third order polynomial equations. Linear models allowed the best fit for data examined at all spatial scales. However, elasticity of the dependent variable to per capita income varied considerably according to scale suggesting that developmental policies may have a limited impact on land quality in vulnerable southern Italian areas compared to northern and central Italy. This study suggests that geographically disaggregated data simulating different spatial levels of governance may offer further insights compared to cross-country datasets indicating targets for multi-scale policies possibly preventing a poverty-desertification spiral.