We analyse the importance of farm level spatial interaction for farm growth. We hypothesize that farms compete on local land markets and interact through knowledge transfer leading to positive and negative feedbacks, respectively. One of the main challenges in the analysis of farm level interaction is to distinguish between actual interactions from the effects of spatially correlated omitted variables. We approach this challenge be estimating a spatially lagged explanatory model (SLX) employing two spatial weighting matrix differentiating between a local and regional neighbourhood. Using a spatially explicit dataset for nearly all Norwegian farms in 1999 and 2009, we found that neighbouring effects differ substantially between local and regional neighbourhood. Our results indicate that the behaviour of directly neighbouring farms is indeed important for farm growth decisions.