Convergence and growth traps in container ports

The aim of this paper is to test for the existence of a middle-rank growth trap in the container port market. This work adopts an original perspective to study the dynamics of port hierarchies by applying different rank-size models, a Markov chain approach and transition modelling to a dataset featuring the annual traffic of 222 container ports from 2000 to 2015. The main findings are that (i) the global port rank-size distribution is characterized by the importance of midsize ports with high traffic volatility; however, the impact of these ports declines over time in favour of the largest ports at the top of the hierarchical system; (ii) port growth does not lead to convergence in port size, except for in the Americas; (iii) there is a relatively low mobility of ports in the container traffic distribution with downward mobility being much higher than upward mobility. These results seem to confirm the existence of a middle-rank growth trap in port hierarchies, which suggests that medium ports should be wary of making large investments to compete with top ports and instead should focus on maintaining their existing position.