During the last year, the research field of spatial economic has rapidly increased. There is consensus that the economic performance of a region depends not only on its own potential, but also on the development of their neighbouring regions. Knowledge spillovers, which are non constant over space, should influence the evolution of the region specific productivity. The so called folk theorem of spatial economics states, that increasing returns to scale are essential for explaining the uneven economic distri- bution of specific economic activity, which implies that knowledge spillover, agglom- eration and distribution of per capita productivity are closely linked. Thus, the aim of this paper is, to introduce a spatial regional growth model, which links first time knowledge spillover, agglomeration, distribution of per capita productivity and the grasp of spillovers. Further, it is shown in a simulation study, how different regimes of returns to scale and grasps of knowledge affect agglomeration and distribution of per capita productivity. One of key findings is, that grasp of knowledge affects dynamic distribution of per capita productivity. Moreover, the simulation study particularly finds support for the folk theorem of spatial economics.