Business performance of airports: Non-aviation revenues and their determinants

• Determinants of non-aviation revenues are investigated empirically.
• Ridge regression and PLSR are used to overcome collinearity of predictors.
• Empirical evidence is based on a sample of German airports.
• Results suggest a potential conflict with the expansion of passengers number.
• Passengers' destination does not seem to be a significant driver.

Management of non-core commercial activities has become a key issue amongst the leverages for improving modern airport industry. Today airports have increased dramatically their dependence on non-aeronautical revenues, which on average account for half of all revenues with this share being highly heterogeneous across regions and airports. Using a dataset of German airports, this paper discusses the improvement of commercial revenues by exploring its determinants. Previous contributions assessed the impact of a selected set of variables non-aviation revenues. Such approach was mainly the effect of multicollinearity, as the majority of relevant variables are strongly correlated to the size of the structures. We address this issue by using ridge regression and partial least squares. Results suggest the potential conflict of non-aviation revenues per passenger and per square meter with the need to expand the number of passengers.