Assessing container operator efficiency with heterogeneous and time-varying production frontiers Jia Yan a, * , Xinyu Sun c , John J. Liu b a School of Economic Sciences, Washington State University, Hulbert 101, Pullman, WA 99164, USA b Faculty of Business, Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, PR China c Faculty of Business Administration, Xi’an University of Technology, PR China article info Article history: Received 16 August 2007 Received in revised form 8 June 2008 Accepted 8 June 2008 Keywords: Stochastic frontier analysis Production efficiency Bayesian approach abstract We build an empirical model under the stochastic frontier framework to assess production efficiencies of container operators from the world’s major container ports in the years between 1997 and 2004. The empirical model measures efficiencies, efficiency changes, and time-persistence of efficiencies after controlling for the individual heterogeneity in technology and technical change. The model is estimated using a Bayesian approach via the Markov Chain Monte-Carlo simulation. We find that the mean efficiency level of the container operators is in the range of 70–90% of their full efficiencies, and the mean effi- ciency changed with time slightly. However, the percentage of highly efficient operators has increased since 1997. Common model misspecifications without controlling for the individual heterogeneity and technical change can alter the results dramatically. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction Since the mid-1970s, most general cargos traditionally transported by the break-bulk method have been transported in containers. The container port industry has become a very important link in the international trade network since then. Although the efficiency of container ports has drawn much attention from both academic research and governmental policy agendas, it is still lacking a rigorous modeling framework that can take the intrinsic characteristics of this industry into ac- count in evaluating efficiency. In this paper, we construct an empirical model for such purpose and measure the technical efficiencies and efficiency changes of container operators from the world’s major container ports in the years between 1997 and 2004, during which major container ports in the world experienced significant institutional changes. 1 One common approach in efficiency studies for container port industry is the use of surveys on subjective perception of port users. Typically, these surveys ask port users to rate a particular port efficiency measure as a scale, for example, a scale from 1 to 7 with 1 indicating the least and 7 indicating the highest efficiency level. An overall efficiency score for a port is then constructed with the ratings on multiple efficiency measures. For example, Sánchez et al. (2003) use this kind of data to construct efficiency measures for Latin American ports and find that such measures are important to explain trade flows. Results obtained from surveys of perceptions are useful for port operators to understand customers’ port choice behavior, but do not necessarily measure the technical efficiency of port production. 0191-2615/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.trb.2008.06.001 * Corresponding author. E-mail address: jiay@wsu.edu (J. Yan). 1 More details on this can be found in Baird (2002). Transportation Research Part B 43 (2009) 172–185 Contents lists available at ScienceDirect Transportation Research Part B journal homepage: www.elsevier.com/locate/trb