Implication of spatial heterogeneity for airports’ efficiency estimation

Dmitry Pavlyuk

Research in Transportation Economics (ISSN: 0739-8859), Vol. 56, pp. 15-24 (2016)
DOI: 10.1016/j.retrec.2016.07.002
Keywords: spatial heterogeneity, airport efficiency, productivity, spatial econometrics, stochastic frontier

Available at ScienceDirect: REQUEST FULL TEXT Export citation: BibTeX RIS


This research is devoted to the analysis of a role of spatial heterogeneity in estimation of airports’ efficiency. Spatial heterogeneity, caused by uneven distribution of influencing factors over space, is widely acknowledged in the European airport industry, but rarely included in airport benchmarking procedures. We utilise modern methods of spatial econometrics to identify the importance of spatial heterogeneity for estimates of airports’ resource elasticity and efficiency values. A set of utilised models includes the spatial error model, spatial stochastic frontier model, geographically weighted regression model, and their modifications.

The empirical findings of the research are based on a sample of 356 European airports (in 2011) and include evidence of the significant effect of spatial heterogeneity on airport’s efficiency and productivity estimates. A multi-output approach to production frontier specification is utilised; served passenger and cargo traffic flows are considered as airport outputs. Elasticities of airport input resources that are infrastructure (runways), catchment area characteristics (population size and welfare), regional tourist intensity, and airport status (hub, international) were estimated and discussed. We paid special attention to the geographical variability of input elasticities over Europe. The paper also contains individual estimates of the sample airports’ efficiency levels.

The main contributions of this paper are the proved necessity of incorporation of spatial heterogeneity into airport benchmarking procedures and the suggested methodological framework for spatial analysis of airports’ productivity and efficiency.