Miranda Gualdrón, Karen AlejandraManjón Antolín, M.Martinez Ibañez, OscarUniversidad Carlos III de Madrid. Departamento de Estadística2019-10-142019-10-142019-10-142387-0303https://hdl.handle.net/10016/29023We present a growth model with spatial interdependencies in the heterogeneous technological progress and the stock of knowledge that, under certain conditions, yields agrowth-initial equation that can be taken to the data. We then use data on EU-NUTS2 regions and a correlated random e ects specication to estimate the resulting spatial Durbin dynamic panel model with spatially weighted individual e ects. QML estimatessupport our model against simpler alternatives that impose a homogeneous technology. Also, our results indicate that rich regions tend to have higher (unobserved) productivityand are likely to stay rich because of the strong time and spatial dependence of the GDP per capita. Poor regions, on the other hand, tend to enjoy productivity spillovers but arelikely to stay poor unless they increase their saving rates.engAtribución-NoComercial-SinDerivadas 3.0 EspañaCorrelated Random EffectsDurbin ModelEconomic GrowthSpatial Panel Dataworking paperC23O47DT/0000001729