Handling Negative Data When Determining Efficiency in the Spanish Hotel Sector

The main purpose of this paper is to study the efficiency of the Spanish hotel sector and its determinants in a two-stage procedure using a profitability approach and a multi-output and multi-input technology. Because profit can adopt negative values, we apply the modified super-efficiency range directional model in the first stage. Then, in the second stage, we use a truncated Bayesian Tobit regression to analyze the effect of a set of determinant variables on the efficiency obtained by non-efficient hotels. Main results show that average efficiency is high in all Spanish regions (around 57.14% of hotels are super-efficient), with several variables such as category or labor productivity having a significant effect on the efficiency achieved by non-efficient hotels.