Efficiency of Brazilian international airports: Applying the bootstrap data envelopment analysis

Abstract

In the setting of open market organizations, efficiency has gained greater relevance, which in Brazil began around the 1990s. This paper applies the bootstrap data envelopment analysis aiming to study the efficiency of Brazilian airports, using the databases of the Agência Nacional de Aviação Civil (ANAC) and the Empresa Brasileira de Infraestrutura Aeroportuária (INFRAERO), dated of 2010, 2011 and 2012. We used multiple regression to validate the variables of the model. In this model, the measure used to represent the performance of airports was the number of processed passengers (dependent variable); for the variables that determine performance (independent variables) we used: number of runways, number of check-in counters, number of aircraft parking bays and passenger areas. The method was applied to study 16 Brazilian international airports in an operational approach. To correct the efficiency values found, given the inherent random error of the data, a bootstrap technique was applied. The approach showed that an airport's size is not the determining factor to assign their efficiency. The use of resources to achieve the product was the most relevant criterion to investigate the airport's good performance and efficiency in this study. The results obtained indicate that the Curitiba Airport is the most efficient. Moreover, the least efficient airports were the Galeão and Manaus Airports.