This paper sets up several random effects spatial autoregressive panel data models with nested Translog production function with non-constant and non-neutral technological change to explain the variation of milk quantity with given stocks of capital, labor (hired and family) and feed production area as other limited productive resource. The aim of the paper is to discover the spatial spillovers in Norwegian milk production as a result of changing quota systems by introducing a spatially lagged variable of milk output. The paper examines three distinctive 5-year balanced panel data sets to account for the evolving milk quota system, such as: no quota, restrictive quota and transferable quota system. The paper also derives joint and conditional Lagrange Multiplier (LM) tests for detecting spatial error correlation (ρ), serial correlation (ψ) and random individual effects (µ) in panel models as well as Moran's I test for testing spatial dependence on panel variables. The tests help to avoid misspecifications of the spatial models. The outcome of the SAREM2SRRE model verified our hypothesis of increasing quota flows within the counties, since the spatial spillover parameters (λ) showed increasing trend under the distinguished dataframes. As a conclusion, the quota system gave rise to positive structural changes because it increased the spatial interdependence and spatial relations between Norwegian dairy farmers.