The aim of this study is to determine the presence or otherwise of transient and/or persistent cost inefficiency in the lodging industry. To do so, we applied the method proposed by Filippini and Greene, within a stochastic frontier panel data framework. Our empirical analysis is based on data for the hotel industry in the Canary Islands (Spain), an important destination for European tourists, for the period 2002-2015. Using a stochastic translog cost frontier model, we show that the industry’s real output during the study period could have been achieved at 17% less cost if persistent (systematic) inefficiency were eliminated, and at 24% less cost in the absence of transient (nonsystematic) inefficiency. We also present evidence of factors that strongly influence hotels’ transient cost efficiency, such as the positive effects of market share (in total revenue terms) and certain management characteristics, including the degree of independence with respect to shareholders, the experience in the industry (years in business) and the number of subsidiaries.

本研究的目的是确定住宿行业是否存在暂时性和/或持续性成本效率低下。为此，我们在随机前沿面板数据框架内应用了 Filippini 和 Greene 提出的方法。我们的实证分析基于 2002 年至 2015 年期间欧洲游客的重要目的地加那利群岛（西班牙）的酒店业数据。使用随机 translog 成本前沿模型，我们表明，如果消除持续（系统性）低效率，研究期间该行业的实际产出可以降低 17% 的成本，而在没有瞬态（非系统性）低效率的情况下，成本可以降低 24% ) 效率低下。我们还提供了强烈影响酒店瞬态成本效率的因素的证据，